How many words do we read per minute?
A review and meta-analysis of reading rate
Department of Experimental Psychology
Keywords: reading rate, reading speed, silent reading, oral reading, language differences, words per
Address: Marc Brysbaert
Department of Experimental Psychology
Tel. +32 9 264 94 25
Based on the analysis of 190 studies (17,887 participants), we estimate that the average silent reading
rate for adults in English is 238 word per minute (wpm) for non-fiction and 260 wpm for fiction. The
difference can be predicted by the length of the words, with longer words in non-fiction than in fiction.
The estimates are lower than the numbers often cited in scientific and popular writings. The reasons for
the overestimates are reviewed. Reading rates are lower for children, old adults, and readers with
English as second language. The reading rates are in line with maximum listening speed and do not
require the assumption of reading-specific language processing. The average oral reading rate (based on
77 studies and 5,965 participants) is 183 wpm. Within each group/task there are reliable individual
differences, which are not yet fully understood. For silent reading of English fiction most adults fall in
the range of 175 to 300 wpm; for fiction the range is 200 to 320 wpm. Reading rates in other languages
can be predicted reasonably well be taking into account the number of words these languages require to
convey the same message as in English.
In a review paper on speed reading, Rayner, Schotter, Masson, Potter, and Treiman (2016) meticulously
explained why reading rates of more than 1,000 words per minute (wpm) are impossible without severe
loss of text understanding. Basically, when we read, we make a sequence of fixations (brief time periods
during which the eyes stand still) and saccades (eye movements to new parts of the text). The text
information we can extract during a fixation is limited and we need time to move our eyes. Both factors
constrain the information that can be extracted from a text in a given time period. Rayner et al. (2016)
were not the first to rebut popular and commercial claims that people can be taught to read much faster
than they usually do without loss of information (e.g., Just & Carpenter, 1987; Spache, 1962; E.A. Taylor,
1957; S.E. Taylor, 1965) and they were not the last (e.g., Seidenberg, 2017).
A question related to the issue of speed reading is how fast we normally read silently. According to
Rayner et al. (2016), for college-educated adults this is “about 200 to 400 wpm” (p. 1). To illustrate the
argument, they presented a table of 10 skilled readers who had an average reading speed of 308 wpm.
The table originally appeared in Rayner (1978) in a review paper on eye movements in reading.
The normal or typical reading rate of 300 wpm is widely mentioned (e.g., Aaron, 2012; Andrews, 2010;
Smith & Pourchot, 1998; Whimbey & Lochhead, 1999; Yaworski, 2005). As it happens, it is one of the
few numbers experimental psychological research has given to society. So, the number is used to
calculate the typical time needed to read online newspaper articles, books, contracts, or legal cases. It is
the speed computer programmers use to present information in rapid successive visual displays (e.g., on
small screens) and it is the number used to determine whether someone is a slow reader (and could
benefit from a remediation program).
In the present article, we discuss how the number came about and how well it is supported by the
The origins of 300 words per minute
The first studies of reading rate in silent reading
To the best of our knowledge, the first author to write about reading rate in a scientific journal was
Quantz (1898). He made a distinction between very slow and very rapid readers. The former had a
reading rate of 3.9 words per second (234 wpm), the latter a rate of 7.3 words per second (438 wpm).
Unfortunately, no information was given about how reading rate had been established. This encouraged
Huey (1901) to reassess the issue. He selected 11 pages from an interesting novel, each containing 405
words, which presented no peculiar difficulties to the reader. Twenty university students were asked to
read one page at a time while Huey measured the time with a stopwatch. There were 10 conditions.
One was normal silent reading (“the way you like to read”). This rate was assessed twice. Another
condition was to read silently as fast as possible, while still being able to follow the story line. Two other
conditions of interest were reading aloud at a normal pace and at a maximal pace.
reading rates of 5.35 and 5.91 words per second for silent normal reading (321 and 355 wpm), 8.21
words per second for silent maximal reading (493 wpm), 3.55 words per second for normal reading
aloud (213 wpm), and 4.58 words per second for maximal reading aloud (275 wpm). The experiment
was included in Huey’s hugely influential book The psychology and pedagogy of reading (Huey, 1908),
together with the data of Quantz (1898). As such, the very first estimates of normal silent reading were
set between 300 and 350 wpm.
An experimental psychologist very active in reading research in the first half of the 20
Tinker. He published a series of over 20 papers addressing various variables that may affect reading
speed, such as letter font and various lay-out options. In most of these studies he used the Chapman-
Cook Speed of Reading test. It had two forms, each containing 30 paragraphs of 30 words. Toward the
end of the paragraph there was an awkward word spoiling the paragraph. Participants had to tick off the
word and finish as many paragraphs as possible in 1.75 minutes. An example of a paragraph was:
“Yesterday I went downtown to buy some shoes and rubbers, but when I got home, I found I had
forgotten to go to the flower-store to get them.”
University undergraduates typically finished some 18 paragraphs (540 words) in 1.75 minutes, making a
reading speed of 309 wpm. In one of the rare studies taking longer than 1.75 minutes per condition,
Tinker (1955) asked his participants to do the task for 30 minutes. They finished on average 317
paragraphs, given a reading rate of 317 wpm (30 words per paragraph and 30 minutes cancelling each
other out). Unfortunately, the task was less than optimal, because the incorrect word was not always
the last word of the paragraph (making the text shorter to read) and participants had to tick off the
errors (taking away some time from the reading). Tinker used easy texts (grade 6 primary school level),
The remaining conditions involved several ways of silently voicing the text, both at normal and maximal pace.
because these were all responded to correctly, so that there was no speed/accuracy issue (Tinker,
1939). Because of his procedure, Tinker rarely mentioned words per minute as dependent variable in his
publications, but he was clear that “even with easy material, 500 wpm is very fast reading” (Tinker,
1958, p. 219).
Eye movement research
In the 1900s it also became feasible to track eye movements in large samples. This motivated a number
of researchers to examine eye movements in normal silent reading and to establish reading norms.
Buswell (1922) published a monograph on eye movements in children and young adults (university
students). The participants had to read a short paragraph of text while their eye movements were
registered. Buswell did not mention reading rate in terms of words per minute, because he found this
too crude a measure (p. 102). Instead, he presented detailed information about the number of fixations
per line of text, the duration of the fixations, and the number of regressive movements per line.
However, he started the monograph with the following sentence: “In the silent reading of an easy
paragraph, Barbara, a first-year pupil, read at a rate of 39.6 words per minute, while Miss. W, a college
Senior, read at a rate of 369 words per minute” (Buswell, 1922, p. 1).
Clearly, for Buswell 300 wpm was a minimum for adult readers. Indeed, in a review paper in 1959 he
summarized the results of research as follows: “The usual rate of reading non-technical material at the
end of the elementary school is about 250 words per minute, while for college students the average is
about 300 words” (Buswell, 1959, p. 113). He went on by saying that the smallness of the increase
beyond the rate of elementary school was a cause of concern, in view of the selective character of the
college population. Buswell believed in the possibility of teaching students to read faster without loss of
comprehension. In his own words: “There have been extreme claims for gains in rate of reading that go
quite beyond the credibility of serious researchers, but there is well substantiated evidence from
research on rate of reading that leaves little room for doubt that a sizable increase in rate without loss in
comprehension could be achieved if schools were to attempt it seriously. There is no support in research
for the popular notion that the slow reader is superior in comprehension. … studies now available
indicate that, at the college level, rate of reading may be forced from 100 to 300 words per minute
above the reader’s present rate without a break in level of tested comprehension” (Buswell, 1959, pp.
Another eye movement researcher, who compared reading performance in children and adults was S.E.
Taylor (1965). Like Buswell, he used eye-movement photography done by special cameras in which light
was reflected from the readers’ eyes and photographed on a moving strip of film. In total, Taylor tested
12,143 readers from first grade through college with at least 1,000 readers per grade. The average
reading rate of the college students was 280 wpm (see below for the rates of the younger readers).
Taylor (1965) did not mention the length of the texts, but given the equipment he used, these cannot
have been longer than one paragraph. In Spichtig, Hiebert, Vorstius, Pascoe, Pearson, and Radach
(2016), the materials were described as five paragraphs of 100 words each.
As indicated above, the number of 300 wpm was also mentioned by Rayner (1978) in his first review
paper on eye movements in reading. It was repeated in the highly cited review paper of Rayner (1998)
and two much used textbooks Rayner co-authored (Rayner & Pollatsek, 1989; Rayner, Pollatsek, Ashby,
& Clifton, 2012). Finally, it figured in the Rayner et al. (2016) paper on speed reading.
A final author influential in promoting the 300 wpm norm was Carver. In a series of publications he
developed a theory of reading inspired by the analogy of a gearbox in a non-automatic car or bicycle.
Just like most cars have five gears with different optimal speeds, Carver ventured that readers used five
reading speeds depending on their reading goal (Caver, 1977, 1982, 1992, 1997). The first reading gear
was memorizing, a situation in which the reader learns a text for free recall. Average reading speed for
this gear was 138 wpm (Carver, 1992). The second reading gear was learning, used when one wanted to
pass a multiple-choice recognition test. Its speed was 200 wpm. The third gear was the one used in
normal silent reading, to understand the text without aiming to answer questions afterwards. Carter
called this gear “rauding" (for reasons explained later) and put it at 300 wpm. The fourth gear was
skimming and used to pick up ideas from a text. Carver put its speed at 450 wpm. Finally, there was a
fifth gear, scanning, which was used to find words in a text. Carver estimated it at 650 wpm.
Carter’s estimate of 300 wpm for normal silent reading was influenced by Buswell and S. E. Taylor, which
he both cited in his 1977 paper, but was also established independently. Before writing his theory of
reading Carver had been involved in the understanding of compressed speech. Gradually, he came to
the conclusion that there was a threshold around 300 wpm, above which the speech suddenly became
much less intelligible.
Information take-up according to Carver involved two opposing forces: the speed with which
information enters the system and the degree to which the information can be picked up by the system.
This trade-off could be investigated by presenting information at different speeds and measuring how
accurate the information take-up was. This was done most prominently in Carver (1982). Passages of
100 words were presented auditorily or visually at presentation rates going from roughly 80 wpm to 500
After the passage, the participants were given cues and had to indicate whether these were
related to the test. They were also asked to estimate the percentage of the passage they had
understood. Efficiency of passage comprehension was then defined as the number of passage thoughts
comprehended per unit of presentation time. As expected, understanding dropped the faster the
information was presented. This was very similar for heard information as for seen information. When
the information presented per time unit was added, information uptake was maximal at 300 wpm, both
for listening and for reading. This number then became Carver’s estimate of the ideal reading rate when
reading for simple information uptake.
Hypotheses of why reading is faster than listening
If reading happens at a rate of 300 wpm, the obvious next question is why reading is faster than spoken
language understanding (which is usually presented at some 150 wpm).
Carver gave one explanation:
Speech remains understandable if it is compressed as long as the compression remains below twice the
normal speed. He coined the word “auding” for this process, defined as listening to words and
determining their meaning. Auding does not involve the actual production of speech and, therefore, can
be faster. In the same way, Carver called looking at words and defining their meaning “rauding” (reading
+ auding). Reading could not be faster for Carver than auding, because the translation of the visual code
to the auditory code was necessary for language understanding. On the basis of existing information,
Carver (1992, p. 89) hypothesized that: “… talking to oneself while operating the rauding process helps
individuals to remember the beginning words of a sentence as the ending words are reached so that the
complete thought can be comprehended”. Already in 1908, Huey had argued that inner speech formed
a central part of silent reading.
Carver had a system of converting word and sentence lengths to standardized measures, which does not concern
For instance, most audiobooks are spoken at a rate of 140-180 wpm (see also Tauroza & Allison, 1990; Yuan,
Liberman, & Cieri, 2006).
A related hypothesis was put forward by Fulford (2001). She ventured that while we are listening to a
person speaking, we simultaneously have an internal conversation preparing to make a response.
Similarly, as a speaker we need an internal conversation because we want to monitor what we are
saying and we may be thinking about making a point stronger. As a result, the total capacity of the
language system is twice the speech rate (300 wpm instead of 150 wpm). The total capacity becomes
available when no response is needed, such as when we are reading a text or listening to an audio tape.
This is why silent reading and auding are fine for speeds up to 300 wpm.
Other researchers hypothesized that the reading rate does not depend on speech-related processes, but
is limited by visual and oculomotor factors. As we saw at the beginning of the article, two aspects are
involved: how much written information can be extracted from the visual field during a fixation, and
how long it takes to move the eyes to a new position. Seidenberg (2017) gave the following rough
• About 7 to 8 letters are read clearly on each fixation.
• Fixation durations average around 200 to 250 milliseconds (4 to 5 per second).
• Words in most texts are about five letters long on average.
• Four fixations per second = 240 fixations per minute.
• 240 fixations × 7 letters per fixation = 1,680 letters per minute
• 1,680 letters/6 (five letters per word plus a space) = 280 words per minute
Some problems with the estimate of 300 wpm
In the previous sections, we saw why authors proposed a normal reading rate of 300 wpm. In general,
they had to defend this number against claims that a little practice was enough to increase the speed to
over 500 wpm. For instance, Fry (1963) claimed that good readers should easily achieve a speed of 350
words per minute, while fair readers reached 250 words, and slow readers 150 words per minute. The
criticism not only came from commercial companies, trying to sell their training programs, but also from
academics arguing that schools and universities should invest in optimizing the reading speed of their
students (e.g., Bellows & Rush, 1952; Buswell, 1959; Deal, 1934; Henry & Lauer, 1939; Jensen, Mills, &
Hershkowitz, 1972; King, Dellande, & Walter, 1969; Maxwell & Mueller, 1965; Poulton, 1961; Stoll, 1974;
Thames & Rosster, 1972; Wooster, 1954).
However, at the same time there were “annoying” findings of normal reading rates well below 300
wpm. A prominent case was the Nelson-Denny Reading Test (Brown, Fishco, & Hanna, 1993; Nelson &
Denny, 1929). In the comprehension subtest, participants are given short text passages of some 200-600
words (drawn from high school and college textbooks) and they have to respond to multiple-choice
questions about the contents of the passages. Students are instructed to read at their normal rate,
neither faster nor slower than usual. Before starting the first passage, they are told that a signal will be
given after one minute and that they have to indicate on the page which word they are reading at that
moment. This is used to calculate the reading rate. There are several versions of the Nelson-Denny test
with their own norms. However, the mean reading rate is typically some 250 wpm and not 300 wpm
(Benevides & Peterson, 2010; Brown et al., 1993; Masterson & Hayes, 2014; Nelson & Denny, 1929).
Another annoying finding came from the type of test used by Tinker. We saw that the Chapman-Cook
Speed of Reading test used by Tinker typically resulted in reading rates around 300 wpm. However,
there were other tests of the same format that gave much lower estimates. The Michigan Speed of
Reading test, for instance, included 100 paragraphs of 30 words and was administered for 10 minutes.
The test takers again had to indicate the awkward words at the end of the passages. The mean number
of paragraphs finished by freshmen for this test was 70.6, equivalent to 212 wpm (Greene, 1934).
Another test of the same construction was the Minnesota Speed of Reading test, which contained 38
paragraphs of on average 52 words with an awkward word towards the end that had to be ticked off.
Students had to complete as many paragraphs as possible in 6 minutes. Eurich & Kraetsch (1982)
mentioned an average reading speed of 17.8 paragraphs (154 wpm) for students tested in 1928 against
15 paragraphs (130 wpm) for students tested in 1978. Importantly, all these numbers are well below
300 wpm. Tinker (1939) argued against the use of these tests because they gave rise to a
speed/accuracy trade-off, suggesting that the awkward words were too difficult to notice in normal
reading. The Michigan and Minnesota tests indeed contained more difficult text paragraphs (college
level), but also took longer than one minute to complete, a characteristic we will return to in the next
Carver’s (1982) finding of an optimal auding rate at 300 wpm was not universally accepted either. After
a major review of the literature, Foulke and Sticht (1969, p. 60) concluded that: “When these studies are
considered collectively, the relationship that emerges is one in which listening comprehension declines
at a slow rate as word rate is increased, until a rate of approximately 275 wpm is reached, and at a
faster rate thereafter” (see also Beatty, Behnke, & Goodyear, 1979). As a matter of fact, Carver set up
his 1982 study to “properly” test the idea of processing thresholds after two findings had been
published that went against his theory of rauding. Jester and Travers (1966) reported different optimal
presentation rates for auditory and visual prose material. Whereas it was 300 wpm for silent reading, it
was 200 wpm for listening. Also Carver himself initially found that nearly every speeding of auditory
information came at a processing cost (Carver, 1973). Not all articles after Carver (1982) have pushed for
an optimal auding rate of 300 wpm either. Rodero (2016), for instance, concluded that the ideal speech
rate for radio news is 170 wpm for high density messages and 190 wpm for low density messages (also
see Rodero, 2012). King and Behnke (1989) reported that comprehension of a formal lecture decreased
as soon as the original signal was compressed. In contrast, short-term memory questions about series of
numbers and letters could be answered well up to 45% compression (i.e., nearly twice the normal rate).
In a third condition, participants had to listen to segments of dialogue and answer questions regarding
the meaning or intentions of the individuals engaged in the dialogue. Here, performance also remained
relatively good up to 45% compression. Unfortunately, no information was given about the speech rate
of the original materials. On the positive side, Conrad (1989) reported that native speakers can repeat
simple sentences speeded up to 320 wpm.
Finally, Rayner (1978) seems to have been selective in the choice of his illustrative table as well. The 10
readers from his table correspond remarkably well to the 10 participants tested in Rayner (1975). In this
study participants read 225 short paragraphs of three to four sentences. They were asked to read the
paragraphs silently while their eye movements were registered. After the readers had read a block of 15
paragraphs, they were shown a set of 12 sentences and asked to identify which of the sentences came
from the passages just read. Interestingly, the 10 participants of Rayner (1975) were described as 10
undergraduate students from Massachusetts Institute of Technology, a university with high entrance
criteria. This may be important because in the same time period Rayner was co-author of two other
papers on text reading (McConkie & Rayner, 1974; McConkie, Rayner, & Wilson, 1973) with much larger
sample sizes and with much lower reading rates (ranging from 190 wpm to 265 wpm). Two reasons
probably convinced Rayner (1998) that the 300 wpm group was more representative. First, it was the
only study with eye movements (the topic of Rayner, 1998). Second, McConkie et al. (1973) and
McConkie and Rayner (1974) showed that participants could be induced to read faster by giving them
awards, without much loss of text comprehension. So, Rayner probably felt right to conclude that
although many of the participants he tested read below 300 wpm, they were easily capable of doing so.
Of course, it is also possible that scientific psychologists are not immune to the social values and general
beliefs in the world around them and are more likely to present data in line with them (Brysbaert &
Rastle, 2013, Chapter 13; Ward, 2002).
A meta-analysis of reading rates
So far we have seen where the idea of 300 wpm as average reading speed came from and why users of
the Nelson-Denny test (widely adopted by schools and universities to detect reading problems) felt
uneasy about the number, given that the Nelson-Denny test has an expected value of 250 wpm. Indeed,
another summary measure often reported for normal reading rate is 250-300 wpm (e.g., Galitz, 2007;
Jonassen, 2004). Because of the discrepancy, a systematic review of the literature is indicated.
There is another reason for a systematic review. Looking at the way reading rate has been assessed, it is
worrying that many tests ask participants to read for one or two minutes only. If we take Carver’s idea of
reading gears seriously, short tests may call for a speed that can be sustained briefly but is not the long
term reading rate. An analogy can be made with the way we move. Arguably, we have two movement
gears: walking and running. In walking there is at least one foot on the ground; in running both feet are
off the ground with each step. Walking does not consume much more energy than resting, can be done
for hours a day, and does not put strain on the body (Carrier, 1984). In contrast, running does consume
extra energy and puts strain on the body. As a result, it tends to be limited to rather short bursts and
requires recuperation afterwards. It is what we do when we are in a hurry or in danger (and what a
small segment of the population likes to do as a workout on a regular basis). Because of the existence of
two types of locomotion we would be puzzled if someone asked us to move forward for 200 meter or
for one minute “in the way we usually move”. Do they mean we should walk or run? Chances are much
lower we would be confused if we were asked to move for ten hours in our usual way, as not many of us
are able to run for that long. Applied to reading, could it be that current tests of reading speed assess
the equivalent of reading as running (meaning that we can do it for a short time after which we are
exhausted and require recuperation) rather than the equivalent of reading as walking (which we can
maintain for most of the day)? If so, we should find faster reading rates for short tests than for long
In summary, we have two questions that can be addressed with a meta-analysis: (1) what is the average
rate of silent reading, and (2) is it faster for short tests than for long tests?
Selection of the studies
We used two methods to find relevant studies. The first was a systematic search based on the Web of
Science, using the selection criteria: ("words per minute" and reading) or "reading rate" or "reading
speed" in Topic. At the last time of testing (February, 2019), this resulted in 2,026 hits. The abstracts
were read and selected if they looked relevant. The criteria were:
- Participants included a group of healthy adults between 17 and 60 years (see below for younger and
- The task involved reading for comprehension or fun.
The criteria resulted in 127 candidates, which were carefully read and pruned. Extra criteria used at this
- The stimulus materials were normal text. This excluded studies where participants responded to
individual words or to random sequences of words without grammatical connections.
- Reading happened silently (see below for oral reading).
- The materials involved a language written in Latin alphabet (see below for other languages).
- At least 10 participants were tested. This excluded psychophysical studies with two or three
- The test was administered to an unselected group of healthy participants. This excluded some
studies in which slow readers were given the opportunity to take part in a reading training program.
For a clinical study to be included, it had to comprise a control condition with healthy participants.
- Participants were native speakers (see below for non-native speakers).
- The task was reading for comprehension or fun. This excluded studies with proofreading or the type
of task Tinker used. It also excluded studies that tested studying for fact retrieval.
- The full text was visible while the participants were reading. This excluded studies in which
participants had to press a button to see the next word (self-paced reading) or studies with various
forms of dynamic presentation (e.g., moving text).
- The article contained enough information to calculate reading rate in words per minute.
The extra pruning left us with 45 relevant articles. These were carefully checked for cross-references,
which pointed us to another 123 articles not covered by the initial search. The studies embrace an era of
more than a century (going from 1901 to 2019) and a large range of topics, including:
- Reading rates for various groups.
- The effects of adverse conditions on reading.
- The effects of various presentation forms on reading rate.
- Eye movement patterns for various types of information.
- Static versus dynamic forms of text presentation.
The vast majority of studies used English stimulus materials, but there were also a few in Dutch, Finnish,
French, German, Italian, Norwegian, Slovenian, Spanish, and Swedish (language differences will be
discussed in a separate section below). Most studies were short, involving a test of a few 100 words or a
test of 1-2 minutes. Some were even shorter and involved unrelated sentences, presented one by one.
Others were longer. The longest study involved five conditions per person with 80 minutes of reading
per condition. It was published by Cushman (1986, Experiment 1) and compared reading on screens and
printed pages. Participants were employees of East Kodak Company. They read articles of general
interest, printed in 10-point Times Roman Medium type with approximately 55 characters per line and
two columns per page. Each article was accompanied by several validated multiple-choice questions for
measuring reading comprehension. In addition to reading rate, measurements of visual fatigue (ocular
discomfort) were taken.
With respect to the short tests, it should be taken into account that often there were multiple
conditions per experiment, meaning that participants in total read for a longer time. As such, they
resemble Huey’s (1901) experiment, in which participants read one page at a time, but did so 11 times.
Similarly, in many eye movement studies, participants were asked to read sentences. However, they
were typically given some 50-100 sentences per experiment.
Nearly all studies involved so-called WEIRD participants (Western, educated, and from industrialized,
rich, and democratic countries;
Henrich, Heine, & Norenzayan, 2010). Participants were predominantly
university students (undergraduates).
As such, the remit of the data is limited to this segment of the
world population. Henry, Van Dyke, and Kuperman (2018) tested a non-university sample in addition to
a university sample on the reading of short texts (see Table 1). Whereas the university students had an
average reading rate of 248 wpm, the mean reading rate of the community sample was 218 wpm.
For each article:
These data are not in the article, but were kindly provided by the authors upon request.
- We selected the most natural condition if there was a choice. This was, for instance, the condition in
which the full text was visible, the condition in which the materials were presented on paper (as we
did for Cushman, 1986)
, or the condition with the longest text to read.
- If no condition stood out, we took the average of all conditions.
- Unless there were clear differences in reading rates for groups in a between-subjects variable, the
data were added and the total number of participants calculated.
All in all, we found 190 studies in 169 articles, involving a total of 17,887 participants. They are
summarized in Table 1 (see supplementary materials for a spreadsheet with more information).
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Insert Table 1 about here
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Because the dependent variable is the same for all studies (wpm), is meaningful and does not involve a
comparison of conditions, we can simply use the dependent variable instead of a standardized measure
of effect size. The average reading rate was 238 wpm (SD = 52.0; 95% confidence interval = 230 – 246).
The median is 235, indicating that the distribution is largely symmetric. When the studies were weighted
for the number of participants they included, the mean increased to 241 wpm, indicating that studies
with large numbers of participants tended to return higher reading rates.
Further analysis of Table 1 did not provide evidence for faster reading rates with short tests than with
long tests, as we had hypothesized. What became evident, was that there was much more variability in
estimates based on short tests than on long tests, very similar to the funnel plot typically observed in
meta-analyses when the effect size is plotted against study precision (Light & Pillemer, 1984). Study
precision is usually expressed in terms of number of participants tested, but for stochastic variables it is
also related to the number of observations taken per participant (Brysbaert & Stevens, 2018; Rouder &
Haaf, 2018). Interestingly, for the present analysis, the funnel plot is even observed for means of
studies, suggesting that the specific conditions (demand characteristics) of each and every study formed
a random variable as well. Apparently, participants in different studies interpret the task differently (e.g.
Reading on paper results in slightly better comprehension than reading on a screen (Clinton, 2019; Delgado,
Vargas, Ackerman, & Salmerón, 2018; Kong, Seo, & Zhai, 2018). Reading times are very similar, except during the
first studies when the quality of the screens was low.
by putting more emphasis on speed or on accuracy) and this is more so for short tests than for long
Figure 1 shows the outcome of the funnel plot analysis, suggesting indeed that there is no need for a
different reading speed in short tests than in long tests. All that needs to be assumed is more variability
in interpretation for studies based on short tests than for studies based on long tests (think of being
asked to move forward for 1 min versus 1 hour “as you would normally move”). Because the funnel plot
is largely symmetric, there are no indications of a publication bias. Interestingly, the funnel plot was less
convincing when the total number of words read in the study was used instead of the number of words
per trial. So, there is less variability in studies that asked participants to read one reasonably long text
for five minutes than in studies that asked participants to read five short texts for one minute each.
Figure 1: Reading rates observed in studies as a function of the number of words read per trial, showing
that studies in which participants read a text for one hour (about 15 thousand words) have less variance
in the estimates than studies in which participants read for one minute or less (about 250 words). Each
dot represents a study from Table 1.
The average reading rate of 238 wpm coming out of the meta-analysis is 20% lower than the
recommended rate of 300 wpm. It is also 5% less than the other recommended reading rate of 250
wpm. However, it could still be explained within Carver’s (1992) theory by assuming that there are two
groups of readers. One group (about 60%) misinterpreted the task as a learning task and read at a speed
of 200 wpm; the other group (40%) rauded as asked. Indeed, nearly all studies asked the participants to
answer (easy) questions to guarantee a minimum of comprehension and the 18 that did not do so, had a
slightly higher reading rate of 265 wpm (see Table 1).
A small book reading study
Fiction reading as a critical test
The only way to be sure that people are rauding as meant by Carver is to observe them while they are
reading a fiction book. Surprisingly, no such study could be found. The five studies in Table 1 that come
closest are the ones by Dwyer and West (1994), Tyrrell, Pasquale, Aten and Francis (2001), Zambarbieri
and Carniglia (2012), Benedetto, Carbone, Drai-Zerbib, Pedrotti and Baccino (2014), and Mak and
Dwyer and West (1994) investigated the effect of sustained silent reading on the reading rate of college
students. A group of 76 students (education majors or students otherwise interested in teaching) were
asked to read novels for 25 slots of 15 minutes each. They were asked to estimate the number of words
read (by counting the words on a few pages and multiplying them with the number of pages read) as
part of a class project in which the effects of sustained reading were investigated. The average reading
rate of the first five blocks (the first week) was 242 wpm. By the last week, the rate had increased to 278
Tyrrell et al. (2001) examined reading performance for three types of display. Eighteen undergraduates
were asked to read the novel Dracula for three session of one hour each: two on LCD displays and one
on a high quality hard copy. There were no differences in reading rates between the conditions and the
average reading speed was 248 wpm.
Zambarbieri and Carniglia (2012) asked 38 participants to read chapters of some 2,000-3,500 words
from a novel in Italian on different devices, while the eye movements were tracked. Reading on average
was 290 wpm, with no big differences between the devices.
Benedetto et al. (2014) asked 48 participants (not further detailed, mean age = 27 years) to read a novel
of Maupassant for an hour. The eye movements were registered but this could be done while the
participants were sitting in a comfortable chair. Between-participant manipulations were the screen
luminance and the ambient illumination. Ambient illumination did not have an effect, but screen
luminance did. The two groups with a high screen luminance (the best condition) had reading rates of
256 wpm and 275 wpm.
Mak and Willems (2018) asked 109 participants to read three short stories (between two thousand and
three thousand words), while their eye movements were tracked. Reading rate was 304 wpm for the
first story, 253 wpm for the second, and 222 wpm for the third.
After each story the participants were
asked three general questions to make sure they had read the stories.
To collect some more data, a study was set up in which regular readers were asked to register the time
they needed to read the next book on their list. Because external validity was more important than
internal validity, participants were recruited via emails to friends and colleagues and via social media.
They were simply asked for their next book to note down the time whenever they started and stopped
reading. After the book was finished, the intervals were added to get the total reading time. The title of
the book was noted together with the total reading time. Participants were between 18 and 60 years
and belonged to the WEIRD group as defined by Henrich et al. (2010). Indeed, 14 of the 48 had a PhD
degree. All others had taken university studies or were taking them.
Several languages were involved: 23 books were read in Dutch, 17 in English, 4 in French, 2 in German, 1
in Italian, and 1 in Hebrew. Sixteen of the non-English books were translations of English books. The
number of words in the books was estimated via a website based on information from Amazon
These data are not in the publication but were kindly provided by the authors upon request. Only data of 38 from
the 43 participants could be retrieved without too much hassle.
These data are not in the publication but were kindly provided by the authors upon request.
The author thanks Heleen Vander Beken and the organization for reading promotion “Iedereen leest” (everybody
reads) for their help with data collection.
(http://wordcounters.com/, consulted in March, 2019) and/or by taking samples from the books and
extrapolating to the complete book. Thirty-two of the books were detective, mystery or fantasy novels.
The 16 other were more highbrow literature.
Results and discussion
Mean number of pages per book was 385 (SD = 166), mean number of words was 107,000 (SD = 65,800).
Reading times ranged from slightly below one hour to over 17 hours. Mean reading rate was 260 wpm
(SD = 87). Figure 2 shows the distribution.
Figure 2: Distribution of reading rates for fiction books
Bayesian analyses indicate that the average reading rate of the book reading study is in line with
the reading rate of 238 wpm obtained in the meta-analysis, although the sample is too small and
the results too diverse to give evidence for the null-hypothesis (BF
= .63; Wagenmakers et al.,
2018; Bayes Factors between 1/3 and 3 do not provide evidence for either the null hypothesis or
the alternative hypothesis). In contrast, there is strong evidence that the obtained reading rate is
different from 300 wpm, even in a non-directional test (BF
To see whether a mixture of two Gaussian distributions (arguably with means around 200 wpm
and 300 wpm) gives a better fit than a single Gaussian distribution (with mean around 260 wpm),
the densityMclust() function of the R package mclust (Scrucca, Fop, Murphy, & Raftery, 2016)
was run. It showed that a model with one Gaussian component is better than a model with 2
components (BIC = -571.4521 vs. -572.7989). The difference in BIC value of 1.35 indicates that
the model with one component is about 3.8 times more likely than the model with two
components (Kass & Wasserman, 1995), which is moderate evidence for the single component
Together with the results from the meta-analysis, the fiction book study points against Carver’s
(1992) proposal of two different reading gears: One with a speed of 200 wpm and one with a
speed of 300 wpm, depending on the reading goal. Both the meta-analysis and the book-reading
study further indicate that the assumption of an average reading rate of 300 wpm deviates from
reality. In the book reading study, only 11/48 participants (23%) had reading rates above 300
wpm, even though the participant pool was an educated sample of very regular readers
many books were so-called page-turners.
Reading rates in the wild
In the previous sections we saw that the typical reading rate is 240 wpm for non-fiction books and 260
wpm for fiction, also when people are reading fiction books for pleasure. Although the new study tried
to make the conditions as non-evaluative as possible, the fact that the participants were measuring their
reading times arguably had the effect that they kept focused on the task.
In principle, reading rates can be studied without pressure to remain on the task. Such observations are
possible in e-readers, digital devices that allow people to read books for pleasure and that keep track of
the reading times (mostly unknown to the reader). Unfortunately, these data are rarely made available
to the public because they infringe on the readers’ privacy.
An exception occurred in 2012 when an e-reader company communicated that the then popular last
book of the Hunger Games trilogy, Mockingjay, was finished by the average reader in seven hours (Alter,
To a large extent the selective nature of the sample is unavoidable, as you can only observe book reading times in
people who read books, just like you can only register marathon times for people who run marathons.
2012). Given that the book has 101,000 words, this estimates the average reading speed for another
page-turner to 241 wpm.
The only other study we could find, happened in Russia (Braslavski, Petras, Likhosherstov, & Gäde,
2016). The authors analyzed 10 months of data from a commercial ebook mobile app. It involved some
three million reading sessions of 8,000 users. The users’ speed showed a normal distribution with a
mean of 150 wpm. Reading speed was fastest for books on cooking and food (161 wpm) and slowest for
poetry (143 wpm). Overall, however, the differences between genres were small and substantially lower
than the 238 wpm based on the meta-analysis.
A first factor involved in the low estimate is that Russian needs some 20% fewer words to express a
message than English (see below). So, reading rate expressed as wpm is likely to be lower in Russian
than in English. If we assume a 4:5 ratio, the equivalent English word rate would be 150/4 * 5 = 187
A second factor is that people in everyday life are less task-focused than when they take part in a study.
At least two elements are involved. First, there is good evidence that people tend to mind wander when
they are reading. Feng, D’Mello, and Graesser (2013) asked participants to read the Nelson-Denny texts
sentence by sentence. Occasionally, after a sentence the participants were asked whether they had
experienced thoughts unrelated to the task during the previous sentence. Participants indicated this was
the case in 42% of the sentences for difficult texts and 36% for easy texts (see also Jackson & Balota,
2012). Reading times increased by 958 ms when participants were mind wandering compared to when
they were on task, and participants were 1.5 times less likely to respond correctly to a comprehension
question related to the sentence.
A second reason why readers may not be reading equally fast when they do not feel monitored is that
their reading may be part of multitasking. Bowman, Levine, Waite, and Gendron (2010) examined the
effects of answering instant messages on reading. Unsurprisingly, they found that participants who
received messages took longer to read an expository text than those who did not. More importantly, a
difference remained when the time needed to read and respond to the messages was subtracted from
the total reading time.
Daniel and Woody (2013) compared the time undergraduates need to read a chapter of an introductory
psychology textbook (unfortunately, no length of the chapter in number of words was given).
Participants were 298 undergraduate students. Goal of the study was to compare printed text to e-
books. Orthogonally, there was a second manipulation. Half of the group read in the laboratory while
being observed. The other half read at home. Students in the latter group were asked to register their
reading time. Students in the lab finished the text in 34 minutes; students reading at home needed 1
hour and 9 minutes. There were no differences in comprehension as assessed by a 30-question
performance quiz. Unfortunately, the two conditions were not completely equal because the group in
the laboratory got the quiz immediately after reading the text, whereas the group reading at home got it
later when they were back in class. Still, Daniel and Woody’s (2013) study is a good reminder that
reading rates observed under controlled conditions are likely to be faster than reading rates observed
under more relaxed circumstances, because unobserved readers rarely remain fully focused on the task.
Life span differences in reading rate
Now that we have a better idea of the normal reading rate in silent reading, we can look at a number of
related issues. The first one is age differences in reading speed.
Reading rates in primary school and secondary school children
There are two large-scale studies looking at how reading rates develop during school years. The first one
is that of S.E. Taylor (1965) mentioned earlier.
S.E. Taylor (1965) tested a minimum of 1,000 readers from grade 1 (first year of primary school) to grade
12 (last year of secondary school), in addition to college students. Each group got stimulus materials
adapted for their level and had to answer at least 70% of the true/false comprehension questions
correctly to be included. Table 2 shows the results. They illustrate the increasing reading speed as
children become older and more practiced in reading. As with many developmental studies, the college
sample is selected more than the other samples, which may explain the considerable difference
between grade-12 students and college students.
Table 2: Taylor’s (1965) data on school-age differences in reading rate
S.E. Taylor’s (1965) study was replicated by Spichtig et al. (2016). They tested 2,203 children from grades
2 – 12 (even grades only). They tried to stay as close as possible to Tayler’s (1965) study (including the
same materials and questions), so that a comparison across 50 years was possible.
Figure 3 shows the results. As can be seen, Taylor’s (1965) finding of an increase in reading rate in
primary and secondary education was confirmed. However, reading rates were substantially lower in
Spichtig et al. (2016) for all but grade 2. In addition, the participants did not have 70% correct on the
yes/no questions in almost one third of the recordings. Spichtig et al. (2016) interpreted these findings
as evidence for a strong decline in word recognition automaticity in the USA between 1960 and 2010. A
comparison with Figure 1 puts this conclusion somewhat in perspective. Large differences are found in
reading rates based on short texts, and Taylor’s (1965) reading rate for college students was well above
the average of the meta-analysis.
More important is the consistent increase in reading rate over age in both studies. Slower reading in
children is characterized by more and longer fixations, shorter forward saccades, and more regressions
to earlier parts of the text, as already noticed by Buswell (1922). Rayner (1986) showed that during
fixations children extract information from a smaller part of the visual field than adults, although he did
not think this was the cause of the slower reading rates. Rather, the higher demands of information
processing made it harder for children to pick up information beyond the currently fixated word. Hence
the need for shorter forward saccades. Using a mathematical model, Reichle, Liversedge, Drieghe,
Blythe, Joseph, White, and Rayner (2013) argued that the typical eye movement pattern seen in children
is mostly due to slower word processing. Indeed, word processing goes faster the more often one has
encountered the words (Brysbaert, Mandera, & Keuleers, 2018; Elgort,
Elgort, Brysbaert, Stevens, & Van
Figure 3: A comparison of Taylor’s (1965) data collected in 1960 and Spichtig et al.’s (2016) data
collected in 2011, both showing the increase in reading rate during primary and secondary school.
Increase in reading rate during college years
The Nelson-Denny norms show an increase of 16 wpm from the first year of college to the fourth year
(Brown et al., 1993). Along the same lines, Masterson and Hayes (2004) reported a difference of 10 wpm
between the first and the third year of undergraduate studies in the United Kingdom. It is not clear to
what extent these increases (also reported by Bear and Imus, 1938) are due to the high reading load at
university or to the continuation of faster reading rates with age in youngsters.
Decrease of reading rate in old age
The first study to investigate silent reading in old age, was Aberson and Bouwhuis (1997). They asked
four groups of five persons from different age categories to silently read 12 easy, entertaining short
stories of 534 words on average. The four age groups were: 36-45 years, 56-65 years, 66-75 years, and
above 75 years. All participants were regular readers with good visual acuity and above average
intelligence. They had graduated from higher education. After each story, participants were asked a
question to ensure comprehension. No significant effect of age was found, in line with the small groups
per age category and the fact that all participants were high performers. At the same time, the oldest
group did not include really good performers (with reading rates above 300 wpm).
Subsequent research has documented processing costs in old age, as could be expected on the basis of
physiological changes as people grow older. Changes in visual abilities occur frequently with old age
(Owsley, 2011). This includes a loss of sensitivity to visual detail and increased suffering from visual
crowding, characterized by reduced ability to recognize visual objects in clutter. Even for old individuals
with good vision, there is evidence that the eye movement pattern is different from young adults (Kliegl,
Grabner, Rolfs, & Engbert, 2004; Rayner, Reichle, Stroud, Williams, & Pollatsek, 2009). Their fixations
tend to be longer, which has been taken as evidence for slower word recognition, and they tend to
compensate for this by adopting a more risky reading strategy. They are more likely to infer the
identities of upcoming words on the basis of prior context. As a result they are more likely than young
adults to skip words and make longer forward saccades. Because the predicted word recognition is not
always correct, older people also show an increase in regressions to previously skipped words. As a
result, contrary to what Aberson and Bouwhuis (1997) concluded on the basis of their small groups of
fluent readers, several authors have reported lower reading rates in old adults than in young adults
(Jackson & Balota, 2012; Kliegl et al., 2004; Rayner et al., 2009; Zang et al., 2016). This drop, however, is
not expected to occur before the age of 60-65 in healthy participants (see also Lott, Schneck,
Haegerström-Portnoy, Brabyn, Gildengorin, & West, 2001, but also see Figure 4 below for reading
aloud). For this reason we included participants up to 60 years in the meta-analysis.
Interestingly, old individuals are also handicapped at understanding compressed speech (
Zion, & Espy-Wilson, 2014)
, suggesting another parallel between auditory and visual language
processing. Finally, it is well-documented that working memory capacity shows an improvement up to
the age of 30 and starts to decrease particularly after the age of 60-70 (Alloway & Alloway, 2013) and
there is evidence that good performance in old age depends on the use of compensation strategies to
counter the decrease in speed of processing (Reuter-Lorenz & Cappell, 2008; Salthouse, 1993;
Morrow, Miller, & Hertzog, 2006
). So, the developmental patterns of reading, listening, and working
memory are quite comparable.
Reading rate and text difficulty
It seems self-evident that reading is faster for easy texts than for difficult texts. Easiness depends on the
demands to understand the text relative to the knowledge and skills of the reader. The demands can be
at the word level, the sentence level, or the text level. Britton, Westbrook, and Holdredge (1978), for
instance, asked participants to read easy and difficult paragraphs. The first two sentences of an easy
paragraph were: “A nobleman and a merchant met in a tavern. For their lunch they ordered soup.” The
first two sentences of a difficult paragraph were: “Sometimes great history is made suddenly and
dramatically. Sometimes it enters our lives on tiptoe, almost warily.” Britton et al. (1978) observed
reading rates of 262 wpm for the easy texts, against 182 wpm for the difficult texts. Comparable findings
were reported by Letson (1959), Oliver, Healy, and Mross (2005), and Conlon and Sanders (2011).
So, ideally a measure of normal reading rate takes into account text difficulty. Miller and Coleman
(1972) noticed that word length correlates very well with text difficulty. If instead of words per minute,
letters per second was used as depend variable, then the effect of text difficulty on reading rate
disappeared (see also Carver, 1976; Coke, 1974).
The average word length for English non-fiction books is 4.6 letters (based on the 650 million word
corpus of Johns & Dye, 2019).
If we use this as the most probable length on which the 238 wpm
estimate from the meta-analysis is based, we can use this value to adapt the expected reading rate to
text difficulty. For instance, the average word length of the two easy text passages given by Britton et al.
(1978) was 4.2 letters. This translates to a difficulty corrected reading rate of 238 * 4.6/4.2 = 261 wpm.
The two most difficult passages had an average word length of 5.4, which translates to a corrected
reading rate of 238 * 4.6/5.4 = 203 wpm. The average word length of fiction books in English is 4.2
letters (based on the two billion word corpus of Johns & Dye, 2019). So, the expected reading rate for
the novel reading study is 238 * 4.6/4.2 = 261 wpm, quite close to the obtained value.
To check whether the differences in reading rate between the three stories of Mak and Willems (2018)
mentioned above, correlated with differences in mean word length, we calculated the word lengths for
the three stories. They were respectively 4.4 letters, 4.5 letters, and 5.1 letters (remember that the
reading rates were 304, 253, and 222 wpm). This is the expected negative correlation. However, the
wpm values do not agree with what would be predicted on the basis of the formula used in the previous
paragraph. This suggests that the formula is specific to English and does not generalize to Dutch, which
I thank Brendan Johns for calculating the average words lengths and kindly making them available.
is a very close language, but on average has longer word lengths (Marian, Bartolotti, Chabal, & Shook,
2012). So, in all likelihood each language will require its own equation for word length correction.
Reading rate for text recall
Carver (1992) argued that reading rate is lower for memorizing texts than for reading, because of the
additional need for rehearsal. He hypothesized that reading rate would drop to 140 wpm if participants
expect recall questions. He referred to a study in which participants after reading had to write down
everything they could remember and in which reading rate dropped below 130 wpm.
A finding that fits very well with Carver’s prediction was published by Kemper and Summer (2001). They
used the Nelson-Denny test, but told the participants that they would have to answer text
comprehension questions without being able to consult the text. This arguably changed the test from
text comprehension to text memory. Under these instructions, Kemper and Summer observed reading
rates of 93 wpm both in young and older adults.
Other evidence in line with Caver’s prediction was published by Greene (1931). He asked students to
study psychology texts of 2,500 words either as fast as possible or as carefully as possible while in
addition taking notes. The former group proceeded at a speed of 212 wpm, the latter at a speed of 104
wpm. A group that read carefully without taking notes had a speed of 122 wpm. Accuracy on
subsequent memory tests was higher in the two slower groups than in the fast group. No additional
advantage was seen for taking notes, rather the reverse, making Greene conclude that taking notes was
a distraction if the notes could not be used for studying.
Comparable figures have been published more recently, although there is a large variability in the study
rates. Rothkopf and Billington (1983) observed that their participants read at a pace of 147 wpm, when
instructed to read the passage carefully and try to remember as much about the passages as they could.
Ackerman and Goldsmith (2011) reported that students needed about 10 minutes to study expository
texts of 1,500 words (150 wpm). Chen and Catrambone (2015) observed that students required on
average 18 minutes to study expository texts of 1,000 words in length (56 wpm), although in this study
participants had to respond three times to metacognitive prompts. A similar number was found by Dirix,
Vander Beken, De Bruyne, Brysbaert, and Duyck (2019): Participants studied at a rate of 54 wpm,
whereas they read matched texts at a rate of 189 wpm. Persky and Hogg (2017) reported that their
students on average needed 3.2 hours to study 7,500 word textbook chapters on physiology and
pharmacokinetics (39 wpm), but studying happened at home (cf. Braslavski et al., 2016). Singer
Trakhman, Alexander, and Silverman (2018) found that students on average needed 10.5 minutes to
study 1,800 word passages from an introductory biology textbook (171 wpm).
Findings less in line with Carver’s hypothesis of a reading pattern specific for memorizing have been
obtained by authors who looked more in detail into what people are doing when studying texts.
Goldman and Saul (1990) reported that the reading patterns could be divided into three strategies. The
most frequent strategy (50%) was one in which participants often made regressions to previous text
parts while reading. The two other strategies were equiprobable (25% each) and consisted either of
reading the text just once, or reading the entire text followed by rereading (parts of) the text.
Participants did not use one strategy consistently, but changed between them for the eight passages
they were asked to study. None of these strategies seemed to match Carver’s (1992) hypothesis of slow
reading because of rehearsal. A peculiarity of Goldman and Saul’s (1990) study was that the participants
saw the text sentence by sentence and had to navigate forward and backward by pressing keys.
Hyona, Lorch, and Kaakinen (2002) measured eye movements while the participants were studying.
They obtained evidence for three big clusters and one small. The biggest cluster (almost half of the
participants) read the text linearly from beginning to end at a speed of 231 wpm. The second largest
cluster (about one quarter) also read linearly but at a much slower speed of 133 wpm. The remaining
participants made many regressions. For the third cluster, these were related to the structure of the text
because the regressions went back to places were new topics had been introduced or summarized. For
the smallest cluster the regressions were nonselective. Both groups with frequent regressions
proceeded at a speed of about 130 wpm. After the study, participants were asked to write a summary of
the text. Best performance (83% correct) was observed for the topic structure processors (the ones with
regressions to the topic heads and summaries). The fast linear readers and the nonselective regressors
performed equally well (77% and 80% correct). Worst performance was for the slow linear readers
(66%). Contrary to Goldman and Saul (1990), the students of Hyona et al. (2002) largely fell in the same
cluster for the two texts they read.
All in all, the more detailed data do not agree well with Carver’s (1992) idea of a separate, slower
reading gear for text memorization due to the need for text rehearsal (see also Dirix et al., 2019; Strukelj
& Niehorster, 2018). The only group that showed this pattern, the slow linear readers of Hyona et al.
(2002), had worst memory performance in addition to being rather uncommon. Instead, what seems to
happen is that text memorizing involves text reading and (if needed) memory structuring. The lower
processing rate is not due to longer fixations and shorter forward saccades, but to relating informative
parts with each other via regressive saccades and rereading. The absence of maintenance rehearsal is in
line with recent doubts about the usefulness of such rehearsal for memory (Lewandowsky & Oberauer,
Individual differences in reading rate
Studies report stable individual differences in reading rate. Reliability of reading rates in the Nelson-
Denny test is .7 (Brown et al., 1993; Kemper & Summer, 2001). It is even higher for longer tests. Ramulu,
Swenor, Jefferys and Rubin (2013) reported a parallel test reliability of r = .95 for two texts with 7,300
words each. Mak and Willems (2018) found an average correlation of .86 between the reading rates for
the three stories they asked their participants to read.
There is little evidence that the differences in reading rate lead to better or worse text comprehension
(Carlson, 1949; Thalberg, 1967). If anything, fast readers tend to be slightly better than slow readers
(e.g., Blommers & Lindquist, 1944; Hebert, 2016, Experiment 1). Arguably, there are at least three
elements involved in the correlation between reading rate and comprehension. First, every reader is
likely to have an optimal language input rate above which comprehension declines, but under which
comprehension also falls because the information comes in too slowly to be integrated into meaningful
chunks (Breznitz & Berman, 2003; Kintsch & van Dijk, 1978). Second, if a slower processing rate is used
to better structure and organize the information via regressions and rereading, this should lead to richer
memories for what was presented (Meyer, Talbot, & Florencio, 1999). Such memories, however, are
only needed in high stake situations, such as studying a syllabus for a demanding exam (or a detailed
memory test devised by a reading researcher), and the extra study time arguably has rapidly diminishing
returns. Finally, people with reading or language difficulties, are likely to have slow reading rates and
low accuracy scores. The first factor predicts a null-correlation between reading rate and
comprehension (because the function is curvilinear); the second factor predicts a negative correlation
(more time on the text predicts a better comprehension score); and the third factor predicts a positive
correlation (the faster the reader the better the comprehension). Because of the three factors involved,
researchers can find very different correlations between reading rate and text comprehension,
depending on the difficulty and the length of the text, the test, and the range of readers investigated.
For the participants of the fiction reading study in Figure 2, we can assume that there is little difference
in comprehension between the fast and the slow readers (or at least none that the readers care about).
Furthermore, we can assume that the individual differences in reading rate are quite stable, given the
length of the texts and the data of Ramulu et al. (2013), although some 50 wpm of the differences could
be due to the difficulty of the books read (see the section on the influence of word length).
An interesting question then is what factors correlate with individual differences in normal reading rate
and, if they correlate, whether they are a cause or consequence of the reading rate.
We do not have
the data (yet) to answer the questions, but we can look at the variables that have been proposed in the
literature. The following is a list of variables that were examined, and references to supporting evidence
(see also the section on reading aloud):
- Speed of visual word decoding (Garcia & Cain, 2014)
- Spoken text comprehension (Hirai, 1999; Jackson & McClelland, 1979)
- Vocabulary knowledge (Dixon, LeFevre, & Twilley, 1988)
- Rapid naming of letters or numbers (Arnell, Joanisse, Klein, Busseri, & Tannock, 2009; Kasperski,
Shany, & Katzir, 2016; Kirby, Georgiou, Martinussen, & Parrila, 2010;
Savage, & Frederickson, 2005
- Letter, name, and word matching (Jackson, 1980; Jackson & McClelland, 1979; Stroud, 1945)
- Short-term memory span (Naveh-Benjamin & Ayres, 1986)
- Working memory span (Baddeley,
Logie, Nimmo-Smith, & Brereton,
1985; Perfetti, 1985)
- Metacognitive knowledge (knowing when your text understanding is good enough for your reading
goal; Jones, Conradi, & Amendum, 2016; Mokhtari & Reichard, 2002)
- Number of book authors known (Choi, Lowder, Ferreira, & Henderson, 2015; Martin-Chang & Gould,
- Auditory word recognition (Breznitz & Berman, 2003)
- Speech rate (
Bosshardt & Fransen, 1996)
Visual acuity (
Aberson and Bouwhuis, 1997)
- Word spelling accuracy (Veldre & Andrews, 2014; Zutell & Rasinski, 1989)
- Intelligence (Hage & Stroud, 1959)
- Speed of finding word associations (Traxler, 1934)
- Amount of reading relative to peers (Choi et al., 2015)
- Processing speed (Choi et al., 2015)
There is a related literature on factors correlating with dyslexia, which is not covered here unless it can be
assumed that the variables also correlate with individual differences in normal reading rate.
Reading rates in non-native speakers
Reading speed in second-language (L2) speakers is considerably slower than in first-language (L1)
speakers. Indeed, reading rates below 100 wpm are no exception. Hirai (1999), for instance, studied
English L2 reading rate in Japanese university students. All students had taken six years of formal English
education in junior and senior high school. In addition, most of the participants had two 90-minute
English lessons per week at the university and a subgroup majoring in English had about five to eight
English courses per week. Text materials were easy prose passages, followed by a set of eight four-
option multiple-choice questions. Reading rate was 139 wpm for participants who could answer more
than 75% of the questions correctly and 61 wpm for the other participants. Interestingly, Hirai (1999)
also tested the participants on English L2 listening and found that their estimated optimal listening rates
corresponded well to the observed reading rates. Before, Conrad (1989) had already observed that
understanding of compressed speech drops much faster in L2 users than in L1 users. This was the case
even for highly proficient L2 speakers who had obtained an average score of 83/100 on the Michigan
State University English Language Exam, which tested the subskills of listening comprehension,
grammar, vocabulary, and writing in English.
Cop, Drieghe, and Duyck (2015) asked reasonably proficient Dutch-English bilinguals to read half a novel
in L1 and the other half in L2. Reading rate was 17% slower in L2 than in L1.
In addition, the eye
movement pattern of L2 readers very much resembled that of L1 children: They made more fixations
per sentence, fixations times were longer, forward saccades were shorter, and less words were skipped.
Only the number of regressions did not differ. Similar results were published by Whitford and Titone
(2014) for sentence reading, although in their study regression rates were higher in L2 than in L1 as well.
Dirix et al. (2019) observed 10% slower processing rates when participants read or studied texts in a
second language (respectively 174 wpm and 50 wpm) than in the first language (189 and 54 wpm).
The similarity of L2 eye movements to children’s eye movements agrees with the hypothesis that a
lower exposure rate to L2 words is the main reason why it takes longer to recognize L2 words than L1
words, particularly for low-frequency words (
Diependaele, Lemhöfer, & Brysbaert, 2013).
Only when L2
readers have the same degree of exposure to L2 words as L1 speakers to L1 words, can we expect both
groups to be equally efficient at reading the language.
Unfortunately, the authors like many other eye movement researchers did not report the statistics needed to
compute reading rate in a way that is comparable to studies without eye movements.
So far, we have discussed silent reading. Another way of testing reading rate is to look at the reading
speed when participants are reading aloud. This has the advantage that the experimenter has more
information (control) of what the participant is doing while reading.
Utility of oral reading rates
Reading aloud is the main way of understanding written text for starting readers, as they need access to
the phonological information in order to understand the words they are reading (Sprenger-Charolles,
Siegel, Béchennec, & Serniclaes, 2003). So, oral reading fluency is a good indicator of reading proficiency
in the first years of primary school (Fuchs, Fuchs, Hosp, & Jenkins, 2001; Rasinski & Hoffman, 2003;
Oral reading fluency is no longer thought to be of importance in higher education, as nearly everyone by
then reads and studies in silence, except maybe in second language education (Gibson, 2008). Still, there
are two research areas in which oral reading rates in adult native speakers are assessed. The first one
has to do with vision problems and their impact on reading. The second one comes from dyslexia
research, where students with dyslexia are compared to healthy controls on reading aloud a paragraph.
Reading charts in ophthalmology
In ophthalmology there is an interesting literature on the development of standardized and normed
reading charts. Most interesting for our research questions are the reading charts consisting of short
paragraphs of text. The most widely normed is the International Reading Speed Texts (IReST; Trauzettel-
Klosinski & Dietz, 2012; see also Radner, Radner, & Diendorfer, 2016, for another set of paragraphs and
their norms in German). The IReST consists of 10 equivalent paragraphs of some 150 words each, and is
available for next to 20 languages (the texts in the different languages are translations of the original,
German texts). The texts have been normed by presenting them to 25 healthy native speakers per
language. The texts are simple (they must be readable by nearly everyone) and the words are short (e.g.,
4.3 letters in the English version). Average reading speed is 228 wpm in English. There are reasons to
believe that the manual administration leads to an overestimate of the true reading rate by some 10%.
This is because the participant can see the chart a few 100 ms before the timer is started (Calabrèse, To,
He, Berkholtz, Rafian, & Legge, 2018).
Other useful charts in ophthalmology present standardized sentences in various font sizes, to see at
what point reading starts to deteriorate. One of these is the Minnesota Low-Vision Reading Test
(MNread; Calabrèse et al., 2016). The test consists of a series of 60-character sentences (some 12-15
words) displayed on three lines. The patient has to read the sentence aloud as fluently as possible.
Figure 4 shows the results as a function of age (the maximal speed refers to the largest font sizes, where
reading speed is optimal). Notice the increase of reading speed below the age of 18 and the decrease in
old age, just like for silent reading (see Hasbrouck & Tindal, 2006, and Ford, Missall, Hosp, & Kuhle,
2017, for more detailed information about reading aloud rates at young ages).
Figure 4: Reading aloud rate as a function of age based on the MNread test (each dot is a different
study; the darker the dot, the more participants were tested)
Source: Calabrèse et al. (2016)
An alternative to the MNread test is the Radner Reading Charts (Radner, Obermayer, Richter-Mueksch,
Willinger, Velikay-Parel, & Eisenwort, 2002), also available for several languages. Each chart of this test
has sentences of 14 words distributed over three lines.
Brussee, van Nispen, and van Rens (2017) compared performance on IRest and the Radner Reading
Charts in Dutch adults. They also looked at the effects of age and education. Figure 5 shows the
outcome. Both tests gave very similar estimates (although the reading rates based on the Radner were
higher than those based on IReST) and showed differences as a function of age and education.
Particularly for the younger participants (18-36 years) there was a strong effect of education.
Figure 5: Comparability of the Radner Reading Charts and the IReST test.
Source: Brussee et al. (2017).
Morrice (2017) examined the importance of instruction. Using the English IReST stimuli, he observed
reading rates of 203 wpm in Canadian students under normal instructions, which stress speed,
compared to 181 wpm when the students were asked to read aloud normally (interestingly, both
reading rates were below the English norms reported by Trauzettel-Klosinski & Dietz, 2012).
A final ophthalmologic study worth discussing at some length, was published by Mackensen and Stichler
(1963). These authors were also interested in normal reading rates for the eye clinic. They tested 622
adult participants both on silent reading and on reading aloud. The texts were excerpts from a novella of
Joseph von Eichendorff, completed in 1823. Time was measured with a stop watch. Education differed in
four categories from academics to uneducated janitors. A further distinction was made between
participants below and above 50 years. Mackensen and Stichler (1963) reported reading aloud times
going from 2.8 seconds per 10 words (214 wpm) for the young academics to 3.85 seconds (156 wpm) for
the old janitors. Silent reading rates varied from 353 wpm to 211 wpm.
The ophthalmological charts are interesting (certainly because they exist for several languages) but
deviate from the texts used in silent reading, because they are considerably shorter and simpler. In
addition, time registration is often manual because the charts are mostly hard copies, so that they can
be presented in a uniform way. A final difference is that the texts are not followed by questions, as is
customary in silent reading studies. All these factors tend to increase the observed reading rates.
Norms for dyslexia tests
A second source of reading aloud rates comes from norming studies of dyslexia test batteries. Many of
these batteries include an oral reading test, because students with dyslexia are known to perform worse
on this test than controls. Callens, Tops, and Brysbaert (2012), for instance, tested 100 Dutch-speaking
healthy undergraduates on silent reading of a text of 1,023 words and a reading aloud text of 582 words.
Both readings were followed by comprehension questions. The authors observed a silent reading rate of
244 wpm and an oral reading rate of 136 wpm in the control participants. As is customary for this type
the oral reading text was more difficult than the silent reading text (words of respectively
5.7 and 4.6 letters), meaning that the reading aloud rate based on this source is likely to be an
The two reading rates of Callens et al. (2012) correlated r = .32 (N = 100, p < .01). One reason for this
rather low correlation could be the difference in text difficulty, as just described. Other contributors may
be the less than optimal reliability of the texts (which was not assessed, as only one measurement per
modality was taken) and the fact that oral and silent reading are influenced by somewhat different
factors. As Callens et al. (2012) investigated a range of variables, we can have a look at the latter
possibility by examining the variables correlating with the reading speeds (Table 3).
Re, Tressoldi, Cornoldi, and Lucangeli (2011), for instance, also used a reading aloud text with words of 5.7
letters in Italian.
Table 3: Significant correlations of silent and oral reading rates with other variables, based on the data
of the 100 control participants of Callens et al. (2012), where more information about the variables can
Variable Wpm_silent Wpm_aloud
Dutch word spelling
Dutch words read out loud correctly in 1 min
English words read out loud correcly in 1 min
Dutch nonwords read out loud correctly in 1 min
Mistakes in Dutch text spelling test
Crystallized IQ (KAIT)
Arithmetic (4 operations)
English word spelling
LASSI preparation for test
Working memory (ordering random sequence of letters)
Fluid IQ (KAIT)
LASSI main idea selection
Spoken text comprehension (KAIT)
Delayed spoken text memory (KAIT)
Time needed to name digits
Silent word reading was influenced more than oral reading by spelling skills (the better the skills the
higher the reading rate), vocabulary and crystalized intelligence (the higher, the higher the reading rate),
and spoken text comprehension. In contrast, reading aloud correlated more with rapid naming of
unrelated stimuli (random words, digits), as could be expected. Interestingly, silent reading rate also
correlated with two non-cognitive factors. Conscientious students read more slowly and students who
thought of themselves as well prepared for tests read more rapidly. The correlation of the two non-
cognitive variables with silent reading but not reading aloud agrees with the hypothesis that students in
a silent reading test feel uncertain about where to put emphasis: speed or accuracy on the test
Lewandowski, Codding, Kleinmann, and Tucker (2003) examined 90 English-speaking students on silent
reading rate (the Nelson-Denny Reading Test) and oral reading rate (reading aloud three 300 word
passages from the Nelson-Denny Reading Test) as part of an assessment battery. Silent reading rate was
231 wpm, oral reading rate was 189 wpm. Both measures correlated .48 with each other. The reliability
of the silent reading rate was not tested (as only one measurement was made), but from the norms we
know it is unlikely to be higher than .7 (Brown et al., 1993). The three oral tests correlated .8 with each
other. The average reading aloud rate correlated more with text comprehension than the silent reading
rate, which could be due to the test’s higher reliability.
Ciuffo et al. (2017) also reported a correlation of r = .48 between silent reading rate and oral reading
rate in Italian speakers. Unfortunately, they did not assess the reliability of their measures.
All in all, we managed to find 77 reading aloud studies (5,965 participants) from 55 articles with
languages based on the Latin alphabet (see Table 4 and supplementary materials). One extra study was
deemed to be an outlier. Kemper, Jackson, Cheung, and Anagnopoulos (1993) reported reading aloud
rates of 295 wpm for college students and 248 wpm for older adults, which seem very unlikely given the
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Insert Table 4 about here
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Mean reading rate was 183 wpm when each study was given equal weight (SD = 25.9, 95% confidence
interval = 177 – 189 wpm). The number was virtually the same when the studies were weighted
according to the number of participants in them (180 wpm). Also the median was virtually the same
(181 wpm), indicating that the distribution was symmetric. Figure 6 shows the findings in a figure with
the same wpm scale as in Figure 1 (silent reading). This figure shows that the reading aloud studies in
general had fewer words per trials, but that this was justified, because the range of observed reading
rates across studies was considerably smaller. A 5 minute test (900 words) or 10 minute test (1,800
words) would seem to guarantee comparable results across studies, certainly when word length is taken
into account (remember we had a combination of easy ophthalmological texts and difficult dyslexia
They would fit perfectly in the picture of silent reading, however.
The average oral reading rate agrees well what can be expected from speaking rates. At the same time,
it is an unrealistic rate for day-to-day use (e.g., someone reading aloud for several hours per day). On
the basis of a study with 10 male adults, Fredericks, Kumar, Oda, and Butt (2015) considered a mean
oral speed of 120 wpm acceptable for an 8-hour day with 40 dB background noise. A similar rate was
recommended by Zagoruiko and Tambovtsev (1982) for experienced operators; for new operators the
recommended rate was lower.
Maximum reading rates
In addition to looking at the average, “normal” reading rate, we can also look at the maximum reading
rates, the equivalents of record speeds in sports. Above, we compared reading to running, which is an
interesting analogy here as well. Thus far, we have mainly talked about the equivalents of long distance
disciplines, involving at least one minute of action.
We can, however, also look at the equivalents of
The record speeds for long-distance running range from 28.5 km/hr for 800 meters (world record = 1 min 41
sec), over 23.7 km/hr for 5 km (world record = 12 min 38 sec), 20.7 km/hr for the marathon (42.2 km, world record
= 2 hr 2 min), to 16.2 km/hr for the 100 km ultra-marathon (6 hr 10 min).
the short disciplines, such as 100 m (9.58 sec = 37.5 km/hr), 200 m (19.19 sec = 37.5 km/hr), or 400 m
(43.03 sec = 33.4 km/hr).
Perrin, Paillé, and Baccino (2015) generated short, easy sentences (in French) of the type “all dogs are
animals”. Participants had to repeat the sentence aloud. Sentence presentation time was decreased
until the participants could no longer do the task. A group of 45 young adults with good vision took part.
The fastest participant needed the sentences to be shown for merely 15 ms, the slowest for 115 ms. This
translates to reading speeds of 16,000 wpm and 2,100 wpm respectively. In all likelihood, these speeds
are only possible in situations when all information can be entered from iconic memory to short-term
memory at once; that is, for sentences containing some 3-4 critical pieces of information (Cowan, 2001;
Rubin and Turano (1992, Experiment 2) used a similar approach to test maximum reading speeds for
longer materials. They used short paragraphs (95 to 125 words in length) and also reduced the
presentation time as long as the participants (middle-aged adults) understood the text. Comprehension
was tested by asking four questions about the text, at least three of which had to be answered correctly.
Rubin and Turano reported a median reading rate of 790 wpm. The fastest participant went up to the
maximum possible speed of 1,652 wpm. This speed was also obtained by six of the nine participants
when the words were presented after each other on the same place (rapid serial visual presentation), so
that the participants did not have to move their eyes, making the authors conclude that eye movements
put a limit to the maximum reading speed in normal reading (see also Primativo, Spinelli, Zoccolotti, De
Luca, & Martelli, 2016)..
It is important to keep in mind that these reading speeds are only possible for very brief time intervals
when maximal use can be made of iconic and short-term memory. In addition, they require a
recuperation period afterwards (just like running 100 meters at maximum speed does). Benedetto et al.
(2015) asked participants to read the first chapter from a novel of Orwell for more than 20 minutes with
rapid serial visual presentation. Participants could press on a pause button when the text was going too
fast. Under these circumstance, participants read at a pace of 200 wpm (very similar to reading a hard
copy text of the same chapter) and performed worse on a comprehension test than participants who
were allowed to read from a book. Similar results were published by Ricciardi and Di Nocera (2017).
The maximum reading rates we discussed are also limited to situations in which participants intended to
read and understand the full text. This is different from situations in which participants skim text for
useful information. For instance, an analysis of newspapers in the Netherlands indicates that each
edition contains some 40 thousand words (more in the weekend; Nederlandse Nieuwsmonitor, 2014).
So, someone reading the newspaper for 1 hour a day, “reads” at a speed of 667 wpm. Someone only
devoting half an hour per day, “reads” at a speed of 1,333 wpm.
As Rayner et al. (2016) remarked, the
maximum reading rate in this situation depends more on the skills for turning pages than on those for
Reading rates in different languages
So far, we have discussed reading rates in Western languages written in Latin alphabet, based on the
assumption that differences between these languages are smaller than the differences between the
studies in each language. At the same time, we have seen that reading rates expressed as wpm depend
on the length of the words (see the difference between fiction and non-fiction).
Differences in average word length and other aspects of the English language may imply that the wpm
estimates discussed so far are to some extent English-specific, given that they are based predominantly
on studies in English.
The following features of English are likely to have an impact on the average word length in the
- English makes extensive use of short function words. These words are limited in number (some 250)
and consist of prepositions (at, between, in, of, …), pronouns (anybody, he, I, it, …), determiners (a,
more, that, the, …), conjunctions (and, because, or, when, …), auxiliary verbs (be, do, get, go, …), and
particles (as, no, not, …). Function words substantially reduce the average word length in English
texts (mean word length of content words in English is 5.9 letters for fiction books and 6.7 letters for
non-fiction books; footnote 7). Languages without (some of the) function words need fewer words
to express the same ideas and, therefore, may have lower reading rates expressed in wpm. For
instance, languages like Chinese, Indonesian, Japanese, Hindi, and Russian do not have articles
The same is true for reading scientific articles. For most purposes, everything worth remembering from this
article (33 thousand words references included) is written in the abstract (194 words). So, everyone reading the
abstract at 238 wpm, (which takes about 50 seconds) “reads” the equivalent of an amazing 40 thousand words per
Various authors have argued that reading rates based letter, syllable or phoneme are better measures than per
words per minute. However, these measures have never caught on beyond the small research communities
involved, as they are more difficult to grasp intuitively. This is why we use wpm in this paper.
before nouns. Other languages use fewer prepositions, because they have case marking or a wider
system of suffixes (e.g., Turkish) or use fewer pronouns (e.g., Arabic).
- English makes a distinction between 44 phonemes (sounds that signal word meanings) whereas
Spanish only distinguishes between 25 phonemes. Cutler, Norris, and Sebastian-Galles (2004)
argued that this difference results in longer spoken words in Spanish than in English. At the same
time, the richness in phonemes in English must be represented with the same set of letters, so that
written syllables in English contain more letters than written syllables in Spanish (Yap, Liow, Jalil, &
- English is written in Latin alphabet and word length depends on how the sounds are represented by
letters of the alphabet. So, words written in other alphabets or in other writing systems can have
radically different lengths. For instance, words in Chinese are much shorter. The vast majority of
words in Chinese texts consist of one or two characters (Chen & Liu, 2014). Indeed, a rule thumb in
the language industry is that a Chinese word on average has 1.5 characters (likely to be more in non-
fiction texts than in fiction texts).
The Chinese language further differs from English because there
are no spaces between the words. Indeed, the concept of words in Chinese is much less prominent
than the concept of character or morphemic unit.
- The English language frequently uses compound words, combinations of two or more words that
function as single units of meaning. The best known are compound nouns (teaspoon, coffee spoon),
but compounding can also involve other parts of speech (to spoon-feed, a spoon-fed student).
Because compounds tend to become long, they are often written as separate or hyphenated words
in English. This is different from other closely related languages such as Dutch or German, which
require compound words to be written as single words. Noun compounding in English goes beyond
what is possible in other languages. For instance, in most languages close to English the expression
“the writer of the text in English” cannot be expressed as “the English text writer”. This drop of
prepositions and articles tends to increase the average word length in English relative to the other
languages. The same is true for the use of the Saxon genitive (“my sister’s daughter’s friend instead
of “the friend of the daughter of my sister”).
Because languages express ideas with different numbers of words of different lengths, it has been
proposed that it is better to look at the information transmitted in a text rather the number of words
Yan, Kliegl, Richter, Nuthmann, and Shu (2010), for instance, reported that the sentences they used from a
Chinese newspaper contained on average 21 characters for 11.2 words, or 1.87 characters per word.
used. Some evidence that reading times are equivalent when text messages are matched, was published
by Liversedge, Drieghe, Li, Yan, Bai, and Hyönä (2016). They wrote eight short texts in English, Finnish,
and Mandarin Chinese. They made sure that the translations were as close as possible. For these stories,
the time needed per sentence to read the information did not differ significantly between the
languages, despite the differences in word length (1.5 characters in Chinese vs. 5.6 letters in English) or
the number of words in the corpus (1,301 in Finnish vs. 1,762 in English). A similar finding was reported
by Kuperman, Siegelman, and Frost (2019). They compared reading times for non-fiction paragraphs
(Wikipedia articles) in Hebrew and English. Again, the times per paragraph did not differ significantly
between the languages, despite the differences in number of words and word lengths.
Still, given the present topic of reading rate, it is worthwhile to see which reading rates expressed as
wpm are observed in different languages, in particular those languages without a Latin writing system.
In total, we managed to find 77 studies from the following languages: Arabic, Chinese, Greek, Hebrew,
Japanese, Korean, Malay, Russian, Turkish, and Urdu. The full dataset is available in the supplementary
materials. Here we give a summary in alphabetic order of the languages for which we managed to find
at least two silent reading studies and two reading aloud studies.
We found 15 studies on Arabic: 10 with silent reading and 5 with reading aloud. Mean silent reading
rate was 181 wpm, mean reading aloud rate 142 wpm.
In Chinese we found 26 studies coming from 23 articles: 18 on silent reading and 8 on reading aloud.
Mean reading rates were respectively 260 wpm and 152 wpm. With respect to these estimates, it is
important to know that we often had to estimate the number of words from the number of characters
given. When we had to do so, we used the conversion 1.5 characters for 1 word. For those texts in which
the words on average were longer than 1.5 characters, our value is an overestimate. One more study
(Yen, Tsai, Chen, Lin, & Chen, 2011) was excluded as an outlier, because according to the authors the
participants read text passages of 2,000 characters in 100 seconds total reading time, giving an
unrealistic speed of 800 wpm.
For Hebrew we found 10 studies: 7 for silent reading and 3 for reading aloud. Of the silent reading
studies, two were considered outliers. Ben-Yehudah and Eshet-Alkalai (2019) examined the effect of
text-highlighting in print and digital reading. They reported a reading rate of 58 wpm in the condition
without highlighting (the rate was even lower in the condition with highlighting). This reading rate is
more in line with what we saw for text studying rather than text reading. A more controversial case is a
study reported by Hanauer (1998). He compared the reading rate for poetry to that for encyclopedic
items. Even though the participants were told that they would be taking part in a reading experiment
and that they should read the texts in their usual manner and in a way that they think is appropriate,
their reading rate for the encyclopedic texts was 42 wpm only. Given the many tasks participants had to
complete after reading based on the text, it is fair to conclude, we think, that this was perceived as
study task rather than a reading task as well. Based on the remaining five studies, average reading rate is
224 wpm for silent reading and 147 wpm for reading aloud.
Finally, for Korean, we were able to locate 7 studies in five articles: 5 with silent reading and 2 with
reading aloud. Silent reading rate was 226 wpm, reading aloud 133 wpm.
Table 5 summarizes the results, together with those for the languages based on Latin alphabet for which
we had at least two independent studies on silent reading and reading aloud. These data allow us to
examine whether there are meaningful, systematic differences between languages.
Table 5: Reading rates for languages for which there are at least two studies for silent reading and
reading aloud. Also given is the expansion/contraction index. This indicates the number of words
needed in other languages to translate a text of 1,000 words in English (based on the translation of
typical texts with Google translate).
Silent Aloud Expansion
Language Nstudies Nparts Wpm Nstudies Nparts Wpm index
Arabic 10 673 181 5 281 142 822
Chinese 18 786 260 4 197 152 980
Dutch 8 407 245 3 161 171 1005
English 144 15409 236 27 3482 190 1000
Finnish 4 96 195 3 204 162 764
French 6 215 214 4 307 180 1062
German 13 853 260 16 2674 169 975
Hebrew 5 168 224 2 83 147 782
Italian 3 253 285 5 511 182 1006
Korean 5 186 226 2 90 133 692
Spanish 6 213 278 6 189 191 1025
Swedish 5 129 218 3 55 163 897
A first variable we can look at is the so-called text expansion/contraction index between languages, the
degree to which texts expand or contract when you translate them from one language to another. This
number is important for translation services and so there are calculators for this.
tend to contradict each other for certain language pairs. In the end we translated the 4 easiest and the 5
most difficult texts of the 36 given by Aquino (1969). These are exemplary for most of the texts used in
reading rate studies. Table 3 gives the resulting estimates for the languages in our dataset.
If we assume
that the information transmitted is more important than the number of words used, then we can
predict that reading rates expressed as wpm will increase in languages requiring more words than
English to express a message, and will decrease in languages that need fewer words. This is what we
find. There is a positive correlation between wpm and the expansion index for silent reading (r = .59, N =
12, p < .05) and for reading aloud (r = .82). If the average of the two reading indexes is used, we get a
correlation of .77.
A second pattern worth checking is the relationship between silent and oral reading rate across
languages. If reading is similar to processing compressed speech, we should not only find a correlation
between both reading rates, but in addition the regression line should have a zero intercept. The data at
hand are not strong enough to draw firm conclusions (the correlation between silent and oral reading
rate is only r = .51), but the pattern looks promising, as can be seen in Figure 7.
Figure 7: Relationship between silent reading rate (abscissa) and oral reading rate (ordinate) across
languages. Each dot represents one language. As expected, there is a positive correlation. More
importantly, the regression line could go through 0, as is needed if reading resembles listening to
An example is https://www.tomedes.com/wordcountratio/helpmyself.
In this paper a quantitative review was made of reading rates. These rates are important for several
reasons. They are needed to decide about deficient reading in various forms (dyslexia, vision deficiency,
slow reading without a clear cause), to gauge the time required for various tasks (most prominently
reading assignments), and to test the quality of new presentation devices and letter fonts. Reading rate
is also important for psychological theories about the reading processes, individual differences in
information processing, language differences, task effects, and metacognition. Finally, the average
reading rate is a value of interest to the public at large, as they often want to compare their own
performance to that of the population. As a result, reading rate is a variable found in many discussions.
Surprisingly, no systematic review of the literature was done yet. Even worse, some of the views
propagated by researchers appear to be based on thin empirical evidence and are not substantiated by
the present analysis. Below we summarize the main conclusions.
Normal silent reading rate in English is 238 wpm for non-fiction and 260 wpm for fiction
The best estimates we have at the moment for silent reading in English are 238 words per minute (wpm)
for non-fiction texts and 260 wpm for fiction texts. There are large and stable interindividual differences,
so that a better summary is a range of 175-300 wpm for silent reading of non-fiction texts and 200-320
wpm for fiction texts. The difference between the two registers can reasonably well be captured by the
fact that longer words are used in non-fiction texts than in fiction texts. To capture word length in a
reading rate calculator for individual texts, the following equation can be used for English: Expected
reading rate text = 238 * 4.6 / average word length text. This will decrease the predicted reading rate for
texts with many long words and increase it for texts without these words. For instance, the average
length of the words in the present article is 5.1 letters, meaning that the expected reading rate is 238 *
4.6/5.1 = 215 wpm (translating to roughly two hours of solid reading for the present article without the
The values observed are considerably below the number of 300 wpm, promoted by various eye
movement researchers (see the introduction). Setting the target reading rate at 300 wpm is unrealistic
for the majority of people and likely to result in disappointment of what can be achieved. On a more
positive note, the reading rates are close to the reading norms of the Nelson-Denny test, suggesting that
rehabilitation services have a better view of what is possible.
The values of 238 and 260 wpm are valid for adults between 18 and 60 years without reading problems.
They are lower for younger children and older adults. In addition, they are only valid for native speakers.
Second language readers have lower reading rates.
Reading rates for other languages can be approximated by looking at the expansion/contraction index
vis-à-vis English (Table 5). Languages requiring more words to convey a message, have a higher reading
rate expressed as wpm, whereas languages requiring fewer words have a lower reading rate. This is
because the amount of information transferred seems to be more important than the number of words
needed to do so.
It may also be good to keep in mind that these are reading rates for people who keep on task the whole
time. Reading rates in less constrained situations are likely to be lower because of mind wandering and
There is no evidence for reading gears except for the distinction between reading and scanning
Carver (1992) proposed the idea of five reading gears: reading for recall (100 wpm), reading for
recognition (200 wpm), reading out of interest (300 wpm), skimming (450 wpm), and scanning (650
wpm). Reading gears can be compared to gears on a bike or a motor, where the sizes of the cogs
determine how much distance is covered per rotation. This is different from increasing the speed of the
rotations (by pedaling faster or pressing the gas pedal). Reading gears can also be compared to the
difference between walking and running, which involve different mechanisms (as opposed to faster or
We failed to find evidence for a distinction between reading for recognition and reading out of interest.
As it happens, when word length was taken into account, both reading rates were very comparable and
in-between the two supposed gears. We found no evidence for a bimodal curve either, which could
have saved the theory.
We further failed to find a different type of reading for recall than for recognition/interest. It is true that
reading for recall takes more time (100 wpm seems to be a good estimate, although in several studies it
was even lower), but this does not seem to be due to a different type of reading. What seems to happen
is an increased structuring and organization of text information, as can be concluded from the many
regressions and rereadings observed. A notion that springs to mind is that of elaborative rehearsal
introduced by Craik and Lockhart (1972). Recalling a text requires it to be well structured in memory.
The best evidence for a difference in gear is between reading and scanning. In the latter condition one
no longer tries to understand the text but to locate a word in the text. In such a situation, forward
saccades are longer and fixations shorter. In addition, fixation durations are much less influenced by the
frequencies of the words (Rayner & Fischer, 1996; Wang, Sui, & White, 2019). Reichle, Pollatsek, and
Rayner (2012) argued that this pattern of eye movements can be understood by assuming that words
are no longer processed for meaning but for form (based on a coarse familiarity check). Using
classification algorithms applied to eye movement data, Simola, Salojärvi and Kojo (2008), and Biedert,
Hees, Dengel and Buscher (2012) reported 60-85% accuracy in deciding whether in a study participants
had read a text for meaning or tried to find particular information in the text.
Whether there is a further distinction between skimming and scanning, as supposed by Carver (1992), is
less clear. Carver saw skimming as a gear to find ideas in a text (proceeding at a rate of 450 wpm),
different from scanning, which was used to find words in a text (at a speed of 650 wpm). It is very well
possible that skimming involves alternations between scanning / skipping (large) parts of text and bouts
of normal reading (when the text looks interesting). In that case, one is likely to find the information if it
falls in a part that is read, and to miss it when not. In addition, we would predict that the distributions of
saccades and fixations are composed of two subdistributions: One for normal reading and one for
scanning. This possibility remains to be tested.
There is no strong evidence for different language processes in reading and listening
In the introduction we saw that researchers proposed hypotheses why silent reading is twice as fast as
auditory language processing. These involved qualitative differences between reading and listening.
They were based on the assumption that silent reading happens at an average rate of 300 wpm.
However, for the reading rates we observed, there is no need to postulate a difference between
auditory and visual text processing. Speech remains understandable to healthy, young participants when
compressed to 260 wpm. Furthermore, participant groups unable to attain such a reading rate (children,
old adults, second language speakers) also seem unable to process spoken language presented at 260
wpm. Finally, a comparison between languages reveals a correlation between oral reading speed and
silent reading speed, in line with the idea that reading resembles the processing of compressed speech
The correspondence between reading and listening can be interpreted in two ways. The first says that it
is because reading depends on auditory word processing. The written text must be translated into a
spoken form before it can be processed. There is indeed good evidence that silent reading involves the
activation of phonology (Frost, 1998; Rastle & Brysbaert, 2006; Shankweiler & Fowler, 2019; Van Orden,,
Johnston, & Hale, 1988), possibly because verbal short term memory relies on a phonological code
(Baddeley, 2012). Alternatively, it could be that the maximum speed in reading and hearing coincides
because both inputs must be translated into an abstract, amodal memory code for thought (Aydede,
2010). The bottleneck then would be the speed with which the abstract memory codes can be build and
The meta-analysis also leads to three recommendations that researchers may want to take to heart. The
first is the observation that many articles do not contain enough information to calculate reading rate.
This was particularly true for eye movement papers (limiting themselves to fixations and saccades, very
much like Buswell, 1922) and correlational studies (where it is surprising to see how often correlations
are reported without descriptives). For every article included in the tables, there were at least two that
could not be included because of insufficient data. In a few cases this could be corrected by contacting
the authors, but most of the time the data were lost irrevocably. If every author in the future reports the
number of words in their texts, the length of the words (in characters and syllables) and the total
reading time (defined as the time between text onset and the end of the reading), a rich database will
build up rapidly at no extra cost. Having this information as part of an article is also very informative,
because it allows readers to see how fast/slow the participants in the study were. This is important
background information to interpret the more detailed findings.
The second recommendation is that a good assessment of reading rate requires more than a single one-
minute test. This is particularly true for silent reading. Although the average of short tests is
representative, there is much more variability between studies for short tests than for long tests
(Figures 1 and 3). On the basis of the present review, a minimum of 5 minutes seems required.
Preferentially, this involves one long text rather than five short ones, as the latter seem more
susceptible to demand characteristics.
Finally, it would be good practice if from now on researchers always reported the reliability of their
variables. Correlations between variables depend on the reliability of the variables, as no variable can
correlate more with another variable than with itself. Reliability can be assessed by looking at the
internal consistency of the data (e.g., split-half correlation or intraclass correlation) or the test-retest
correlation (e.g., Revelle & Condon, 2018).
The meta-analyses reported in the present article have been able to settle a number of important
issues. At the same time, the analyses make clear that there are still important gaps in our knowledge. In
particular, it is not clear whether the patterns observed at the macro-level (between tasks, languages)
are also valid at the meso-level (predicting reading rates for specific texts, tasks, participants) and at the
micro-level (predicting processing times at the level of individual words). Below a few remaining issues
Is the relationship between silent and oral reading observed for individuals?
If silent reading resembles listening to compressed speech, then the relationship found in Figure 6
(including the zero intercept) should also be observed when individuals read texts silently or orally.
Furthermore, we can expect the relationship to be present both when the intended speed is maximal or
the individual’s long-term average. The only factor that is expected to change is the slope of the
regression line (indicating the degree of compression). Finally, we should see similar decreases in
comprehension when speech and visual text are presented at rates higher than the individual’s reading
speed. Important for this research is that matched stimulus materials are used in visual and auditory
modality, something not taken into account so far.
Are languages equally good at conveying information?
We saw two studies that compared languages and reported that languages seem to be equivalent at
conveying information despite differences in word lengths and writing systems (Kuperman et al., 2019;
Liversedge et al., 2016). This is important information, but we must take into account that both studies
were severely underpowered to test realistic differences in a between-groups design. Liversedge et al.
only had 20-25 participants per group and Kuperman et al. tested 56 participants per group. Such small
designs cannot reliably pick up effect sizes of d = .4, which seems to be the mean effect size in
psychology (Stanley, Carter, & Doucouliagos, 2018). For such an effect size we require at least 100
individuals per language.
The issue is all the more important because Shimron and Sivan (1994) reported slower reading times for
Hebrew texts than for matched English texts, a trend that was also present in Kuperman et al. (2019).
So, there are reasons to look at this issue in good detail.
How well can we predict individual differences in normal adult reading rates?
There are stable differences in reading rate between healthy adults. However, little research has
examined how well these can be predicted and which variables are important.
A seminal series of studies on the topic was run by Jackson and McClelland and published in a number of
papers (Jackson, 1980; Jackson & McClelland, 1975, 1979). They asked university students to read a
short text of 4,286 words and to answer 10 comprehension questions. On the basis of this test, Jackson
and McClelland (1975) selected a group of six average readers (200-300 wpm) and six fast readers (more
than 450 wpm) with equal comprehension. The participants then took part in a series of studies to
determine the differences between them.
The tasks involved brief (tachistoscopic) presentation of the
following stimuli, which the participants had to report:
- Five-word sentences of the type “Dan fixed the flat tire”.
- Letter and letter-pair identification
- Identification of a series of 8 unrelated consonants
- Discriminating between two very similar words
The fast group outperformed the average group on all tasks, except for the letter and the letter-pair
identification tasks, making Jackson and McClelland (1975) conclude that faster readers were capable of
encoding more information for higher-level conceptual processes from each fixation.
Jackson and McClelland (1979) selected a group of 12 slow readers and a group of 12 fast readers and
presented them again with a series of tasks. They measured performance by correcting reading rate for
the percentage questions answered correctly. Three tasks accounted for nearly all systematic variance
between the groups: Listening comprehension, letter name matching (a and A have the same letter
name) and a homophone decision task (doe and dough sound the same). At the same time, the authors
observed that the fast readers made more errors in the experimental tasks, suggesting that they had a
lower accuracy criterion. Visual discrimination did not make a difference between the groups.
Jackson (1980) extended the finding that fast readers are better at matching meaningful stimuli. He
showed that fast readers more rapidly decided whether two pictures belonged to the same category
(e.g., a picture of a dog and a chicken). He also taught them names for meaningless novel characters and
found that fast readers were better at deciding if two characters had the same name or not. At the same
time, fast readers were not faster at indicating whether two meaningless, novel characters were
physically the same or not.
Palmer, MacLeod, Hunt, and Davidson (1985) also examined which aspects of information processing
correlate with normal adult reading. They replicated the high correlation between reading
comprehension and listening comprehension, but on the basis of their findings concluded that reading
rate was a distinct ability, only moderately correlated with reading comprehension. Reading speed
correlated with visual word processing skills, as measured with a word search task (deciding whether a
target word was in a set or not) and a word matching task (deciding that sink SINK are the same word
Notice the low power of the study.
but wink SINK not). Interestingly, the same tasks with letters had much lower correlations with reading
Nearly all articles after Palmer et al. (1985) have focused on the reading comprehension part, meaning
that the reading rate topic has remained largely untouched. In particular, we have no idea how well the
various variables proposed predict individual differences in reading rate.
Are there better measures to predict reading speed for individual texts?
Word length and the expansion / contraction index seem to work surprisingly well to account for
differences in reading rate at the macro level (differences between fiction and non-fiction, between
reading rates of languages). How well do they work to predict the reading rate of individual texts? Above
we saw already that the equation connecting reading rate to word length is likely to be language
specific, depending on the average length of words in the language. In addition, word length is only one
of the (less important) variables explaining text and word difficulty (Chen & Meurers, 2016; Crossley,
Skalicky, Dascalu, McNamara, & Kyle, 2017; Dirix, Brysbaert, & Duyck, 2019; Vajjala, & Meurers, 2014).
So, a remaining, important research question is how much the other variables (e.g., average sentence
length, word frequency, …) help to predict reading rates of individual books on top of average word
length. This has theoretical implications but also could improve the calculations of expected reading
times for individual texts.
Most data discussed in the present review are available as supplementary materials on the website of
the article. They are also available on the open science framework website https://osf.io/3wfas/.
The author thanks the Faculty of Psychology and Educational Sciences at the University of Ghent and
Antje Meyer at the Max Planck Institute for Psycholinguistics in Nijmegen for giving him the opportunity
to finally focus for 2 months on the writing of a ms that had a gestation time of over 5 years and got an
extra urgency after the publication of Rayner et al. (2016).
* means that the article is part of the meta-analyses
Aaron, P. G. (2012). Dyslexia and hyperlexia: Diagnosis and management of developmental reading
disabilities. Springer Science & Business Media.
Aberson, D. H., & Bouwhuis, D. G. (1997).* Silent reading as determined by age and visual acuity. Journal
of Research in Reading, 20(3), 184-204.
Ackerman, R., & Goldsmith, M. (2011). Metacognitive regulation of text learning: On screen versus on
paper. Journal of Experimental Psychology: Applied, 17(1), 18-32.
Acklin, D., & Papesh, M. H. (2017).* Modern speed-reading apps do not foster reading comprehension.
American Journal of Psychology, 130(2), 183-199.
Ahissar, M. (n.d.).* Public Domain Hebrew Reading Data. Available at
Akhmadeeva, L., Tukhvatullin, I., & Veytsman, B. (2012).* Do serifs help in comprehension of printed
text? An experiment with Cyrillic readers. Vision Research, 65, 21-24.
Alabdulkader, B., & Leat, S. J. (2017).* Toward developing a standardized Arabic continuous text reading
chart. Journal of Optometry, 10(2), 84-94.
Alarfaj, A., & Alshumaimeri, Y. (2012).* The effect of a suggested training program on reading speed and
comprehension of Saudi female university students. Procedia-Social and Behavioral Sciences, 31, 612-
Alió, J. L., Radner, W., Plaza-Puche, A. B., Ortiz, D., Neipp, M. C., Quiles, M. J., & Rodríguez-Marín, J.
(2008).* Design of short Spanish sentences for measuring reading performance: Radner-Vissum test.
Journal of Cataract & Refractive Surgery, 34(4), 638-642.
Alloway, T.P. & Alloway, R.G. (2013) Working memory across the lifespan: A cross-sectional approach,
Journal of Cognitive Psychology, 25(1), 84-93, DOI: 10.1080/20445911.2012.748027
Almuhajri, M. (2013).* Arabic E-Reading: Studies on Legibility and Readability for Personal Digital
Assistants. Doctoral dissertation, Concordia University. Available at
Alotaibi, A. Z. (2007).* The effect of font size and type on reading performance with Arabic words in
normally sighted and simulated cataract subjects. Clinical and Experimental Optometry, 90(3), 203-206.
Altarriba, J., Kroll, J. F., Sholl, A., & Rayner, K. (1996).* The influence of lexical and conceptual constraints
on reading mixed-language sentences: Evidence from eye fixations and naming times. Memory &
Cognition, 24(4), 477-492.
Alter, A. (2012). Your E-book is reading you. Wall Street Journal. Available at
Ambrosino, R. J., Brading, P. L., & Noval, L. K. (1974).* Reading: A potential source of academic difficulty
in medical school. Journal of Reading Behavior, 6(4), 367-373.
Andrews, S. (Ed.). (2010). From inkmarks to ideas: Current issues in lexical processing. Psychology Press.
Aquino, M. R. (1969). The validity of the Miller-Coleman readability scale. Reading Research Quarterly,
Arnell, K. M., Joanisse, M. F., Klein, R. M., Busseri, M. A., & Tannock, R. (2009).* Decomposing the
relation between Rapid Automatized Naming (RAN) and reading ability. Canadian Journal of
Experimental Psychology/Revue Canadienne de Psychologie Expérimentale, 63(3), 173-184.
Ashby, J., Yang, J., Evans, K. H., & Rayner, K. (2012).* Eye movements and the perceptual span in silent
and oral reading. Attention, Perception, & Psychophysics, 74(4), 634-640.
Awadh, F. H., Phénix, T., Antzaka, A., Lallier, M., Carreiras, M., & Valdois, S. (2016).* Cross-language
modulation of visual attention span: an Arabic-French-Spanish comparison in skilled adult readers.
Frontiers in Psychology, 7, 307.
Aydede, M. (2010). Language of thought. In Stanford Encyclopedia of Philosophy. Available at
Azmi, A. M., & Alsaiari, A. (2014).* A calligraphic based scheme to justify Arabic text improving
readability and comprehension. Computers in Human Behavior, 39, 177-186.
Baddeley, A. (2012). Working memory: theories, models, and controversies. Annual Review of
Psychology, 63, 1-29.
Baddeley, A., Logie, R., Nimmo-Smith, I., & Brereton, N. (1985).* Components of fluent reading. Journal
of Memory and Language, 24(1), 119-131.
Bai, X., Yan, G., Liversedge, S. P., Zang, C., & Rayner, K. (2008).* Reading spaced and unspaced Chinese
text: Evidence from eye movements. Journal of Experimental Psychology: Human Perception and
Performance, 34(5), 1277-1287.
Baker, R. (2006).* The effects of multiple column online text on reading speed, reading comprehension,
and satisfaction. PhD Thesis Wichita State University. Available at
Ball, R., & Hourcade, J. P. (2011).* Rethinking reading for age from paper and computers. International
Journal of Human-Computer Interaction, 27(11), 1066-1082.
Bassin, C. B., & Martin, C. J. (1976).* Effect of three types of redundancy reduction on comprehension,
reading rate, and reading time of English prose. Journal of Educational Psychology, 68(5), 649-652.
Bear, R. M., & Imus, H. A. (1939).* Changes in reading performance during the freshman year of college.
Journal of Educational Psychology, 30(9), 667-673.
Beatty, M. J., Behnke, R. R., & Goodyear, F. H. (1979). Effects of speeded speech presentations on
confidence-weighted and traditional comprehension scores. Communications Monographs, 46(2), 147-
Bell, L. C., & Perfetti, C. A. (1994).* Reading skill: Some adult comparisons. Journal of Educational
Psychology, 86(2), 244-255.
Bell, S. M., Miller, K. C., McCallum, R. S., Hopkins, M., & Hilton-Prillhart, A. (2012).* Unique screener of
reading fluency and comprehension for adolescents and adults. Psychology, 3(1), 45-48.
Bellows, C. S., & Rush Jr, C. H. (1952).* Reading abilities of business executives. Journal of Applied
Psychology, 36(1), 1-4.
Benedetto, S., Carbone, A., Drai-Zerbib, V., Pedrotti, M., & Baccino, T. (2014).* Effects of luminance and
illuminance on visual fatigue and arousal during digital reading. Computers in Human Behavior, 41, 112-
Benedetto, S., Carbone, A., Pedrotti, M., Le Fevre, K., Bey, L. A. Y., & Baccino, T. (2015).* Rapid serial
visual presentation in reading: The case of Spritz. Computers in Human Behavior, 45, 352-358.
Benevides, T., & Peterson, S. S. (2010).* Literacy attitudes, habits and achievements of future teachers.
Journal of Education for Teaching, 36(3), 291-302.
Benton, S. L., Kraft, R. G., Glover, J. A., & Plake, B. S. (1984).* Cognitive capacity differences among
writers. Journal of Educational Psychology, 76(5), 820-834.
Ben-Yehudah, G. & Eshet-Alkalai, Y. (2018).* The contribution of text-highlighting to comprehension: A
comparison of print and digital reading. Journal of Educational Multimedia and Hypermedia, 27(2), 153-
Berger, A. (1966).* Effectiveness of four methods of increasing reading rate, comprehension, and
flexibility. Research Report, Syracuse University. Available at
Biedert, R., Hees, J., Dengel, A., & Buscher, G. (2012, March). A robust realtime reading-skimming
classifier. In Proceedings of the Symposium on Eye Tracking Research and Applications (pp. 123-130).
Available at http://joernhees.de/dump/papers/2012ETRASkimmingClassifier.pdf.
Birkmire, D. P. (1985).* Text processing: The influence of text structure, background knowledge, and
purpose. Reading Research Quarterly, 20(3), 314-326.
Blommers, P., & Lindquist, E. F. (1944). Rate of comprehension of reading; its measurement and its
relation to comprehension. Journal of Educational Psychology, 35(8), 449-473.
Boije, S. S., & Gustafsson, M. (2012).* Evaluation of reading eye movements and comfort on different
paper design. Master Thesis, Karolinska Institute. Available at http://whitelines.se/wp-
Bosshardt, H. G., & Fransen, H. (1996). Online sentence processing in adults who stutter and adults who
do not stutter. Journal of Speech, Language, and Hearing Research, 39(4), 785-797.
Booker, I. A. (1934).* Measurement and improvement of silent reading among college freshmen. Ph.D.
Dissertation, University of Chicago.
Bowers, A. R., & Reid, V. M. (1997).* Eye movements and reading with simulated visual impairment.
Ophthalmic and Physiological Optics, 17(5), 392-402.
Bowman, L. L., Levine, L. E., Waite, B. M., & Gendron, M. (2010).* Can students really multitask? An
experimental study of instant messaging while reading. Computers & Education, 54(4), 927-931.
Braslavski, P., Petras, V., Likhosherstov, V., & Gäde, M. (2016, October). Ten months of digital reading:
An exploratory log study. In N. Fuhr, L. Kovacs, T. Risse, & W. Nejdl (Eds.), Research and advanced
technology for digital libraries (pp. 392-397). Springer.
Breznitz, Z., & Berman, L. (2003). The underlying factors of word reading rate. Educational Psychology
Review, 15(3), 247-265.
Bridgeman, B., & Montegut, M. (1993, September).* Faster flicker rate increases reading speed on CRTs.
In Human Vision, Visual Processing, and Digital Display IV (Vol. 1913, pp. 134-146). International Society
for Optics and Photonics.
Britton, B. K., Westbrook, R. D., & Holdredge, T. S. (1978).* Reading and cognitive capacity usage: Effects
of text difficulty. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 582-591.
Brown, J.A., Fishco, V.V., & Hanna, G. (1993). Nelson–Denny Reading Test: Manual for Scoring and
Interpretation, Forms G & H. Rolling Meadows, IL: Riverside Publishing.
Brubaker, R. S. (1972).* Rate and pause characteristics of oral reading. Journal of Psycholinguistic
Research, 1(2), 141-147.
Brussee, T., van Nispen, R. M., & van Rens, G. H. (2017).* Visual and personal characteristics are
associated with reading performance in normally sighted adults. Clinical and Experimental Optometry,
Brysbaert, M., Mandera, P., & Keuleers, E. (2018). The word frequency effect in word processing: An
updated review. Current Directions in Psychological Science, 27, 45-50.
Brysbaert, M., & Rastle, K. (2013). Historical and conceptual issues in psychology (2
UK: Pearson Education.
Brysbaert, M. and Stevens, M. (2018). Power Analysis and Effect Size in Mixed Effects Models: A Tutorial.
Journal of Cognition, 1: 9, 1–20, DOI: https://doi.org/10.5334/joc.10.
Buari, N. H. & Chen, A.H. (2017).* Buari-Chen Malay Reading Chart (BCMRC): Contextual Sentence and
Random Words 2-in-1 Design in Malay Language. Pertanika Journal of Science and Technology, 25, 135-
Buari, N. H., Chen, A. H., & Musa, N. (2014).* Comparison of reading speed with 3 different log-scaled
reading charts. Journal of Optometry, 7(4), 210-216.
Buari, N. H., Yusuf, N. H., Satali, A. M., & Chen, A. H. (2015).* Repeatability of the Universiti Teknologi
MARA reading charts. Bangladesh Journal of Medical Science, 14(3), 236-240.
Burt, J. S., & Fury, M. B. (2000).* Spelling in adults: The role of reading skills and experience. Reading
and Writing, 13(1-2), 1-30.
Buswell, G.T. (1922). Fundamental reading habits: a study of their development. University of Chicago:
Supplementary Educational Monographs.
Buswell, G. T. (1959). The process of reading. The Reading Teacher, 13(2), 108-114.
Calabrèse, A., Cheong, A. M., Cheung, S. H., He, Y., Kwon, M., Mansfield, J. S., ... & Legge, G. E. (2016).*
Baseline MNREAD measures for normally sighted subjects from childhood to old age. Investigative
Ophthalmology & Visual Science, 57(8), 3836-3843.
Calabrèse, A., To, L., He, Y., Berkholtz, E., Rafian, P., & Legge, G. E. (2018).* Comparing performance on
the MNREAD iPad application with the MNREAD acuity chart. Journal of Vision, 18(1), 8-8.
Callens, M., Tops, W., & Brysbaert, M. (2012).* Cognitive Profile of Students Who Enter Higher
Education with an Indication of Dyslexia. PLoS ONE 7(6): e38081. doi:10.1371/journal.pone.0038081.
Calossi, A., Boccardo, L., Fossetti, A., & Radner, W. (2014).* Design of short Italian sentences to assess
near vision performance. Journal of Optometry, 7(4), 203-209.
Calvo, M. G., Eysenck, M. W., Ramos, P. M., & Jiménez, A. (1994).* Compensatory reading strategies in
test anxiety. Anxiety, Stress, and Coping, 7(2), 99-116.
Campbell, A. J., Marchetti, F. M., & Mewhort, D. J. K. (1981).* Reading speed and text production A note
on right-justification techniques. Ergonomics, 24(8), 633-640.
Carlson, T. R. (1949). The relationship between speed and accuracy of comprehension. The Journal of
Educational Research, 42(7), 500-512.
Carrier, D.R. (1984). The energetic paradox of human running and hominid evolution. Current
Anthropology, 25(4), 483-495.
Carver, R. P. (1970).* Effect of a "chunked" typography on reading rate and comprehension. Journal of
Applied Psychology, 54(3), 288-296.
Carver, R. P. (1973). Effects of increasing the rate of speech presentation upon comprehension. Journal
of Educational Psychology, 65(1), 118-126.
Carver, R.P. (1976). Word length, prose difficulty, and reading rate. Journal of Reading Behavior, 8(2),
Carver, R. P. (1977). Toward a theory of reading comprehension and rauding. Reading Research
Quarterly, 13(1), 8-63.
Carver, R. P. (1982). Optimal rate of reading prose. Reading Research Quarterly, 18(1), 56-88.
Carver, R. P. (1983).* Is reading rate constant or flexible? Reading Research Quarterly, 18(2), 190-215.
Carver, R. P. (1992). Reading rate: Theory, research, and practical implications. Journal of Reading, 36(2),
Carver, R. P. (1997). Reading for one second, one minute, or one year from the perspective of rauding
theory. Scientific Studies of Reading, 1(1), 3-43.
Cavalli, E. (2016).* La lecture de l’adulte dyslexique universitaire : dépistage, réorganisation cognitive et
neuro-fonctionnelle. PhD Thesis, Université Aix-Marseille.
Cavalli, E., Colé, P., Leloup, G., Poracchia-George, F., Sprenger-Charolles, L., & El Ahmadi, A. (2018).*
Screening for Dyslexia in French-Speaking University Students: An Evaluation of the Detection Accuracy
of the Alouette Test. Journal of Learning Disabilities, 51(3), 268-282.
Chamorro, Y., Treviño, M., & Matute, E. (2017).* Educational and cognitive predictors of pro-and
antisaccadic performance. Frontiers in Psychology, 8, 2009.
Chan, A. H. S., & Lee, P. S. K. (2005).* Effect of display factors on Chinese reading times, comprehension
scores and preferences. Behaviour & Information Technology, 24(2), 81-91.
Chen, A. H., Buari, N. H., & Jufri, S. (2017).* The Development of SAH Reading Passage Compendium: A
Tool for the Assessment of Reading Performance Related to Visual Function. International Education
Studies, 10(12), 30-37.
Chen, D. W., & Catrambone, R. (2015, September). Paper vs. screen: Effects on reading comprehension,
metacognition, and reader behavior. In Proceedings of the human factors and ergonomics society annual
meeting (Vol. 59, No. 1, pp. 332-336). Sage CA: Los Angeles, CA: SAGE Publications.
Chen, H., & Liu, H. (2014). A diachronic study of Chinese word length distribution. Glottometrics, 29, 81-
Chen, X., & Meurers, D. (2016). Characterizing text difficulty with word frequencies. In Proceedings of
the 11th Workshop on Innovative Use of NLP for Building Educational Applications (pp. 84-94).
Choi, W., Lowder, M. W., Ferreira, F., & Henderson, J. M. (2015).* Individual differences in the
perceptual span during reading: Evidence from the moving window technique. Attention, Perception, &
Psychophysics, 77(7), 2463-2475.
Ciuffo, M., Myers, J., Ingrassia, M., Milanese, A., Venuti, M., Alquino, A., ... & Gagliano, A. (2017).* How
fast can we read in the mind? Developmental trajectories of silent reading fluency. Reading and Writing,
Cline, R. K. (1969).* Reading ability and selection for teacher education programs. Journal of Reading,
Clinton, V. (2019). Reading from paper compared to screens: A systematic review and meta-analysis.
Journal of Research in Reading. Advance publication available at https://doi.org/10.1111/1467-
Cohen, J. M., & Waiss, B. (1991).* Comparison of reading speed in normal observers through different
forms of equivalent power low vision devices. Optometry and Vision Science, 68(2), 127-131.
Coke, E. U. (1974). The effects of readability on oral and silent reading rates. Journal of Educational
Psychology, 66(3), 406-409.
Conlon, E., & Sanders, M. (2011).* The reading rate and comprehension of adults with impaired reading
skills or visual discomfort. Journal of Research in Reading, 34(2), 193-214.
Conrad, L. (1989). The effects of time-compressed speech on native and EFL listening comprehension.
Studies in Second Language Acquisition, 11(1), 1-16.
Cop, U., Drieghe, D., & Duyck, W. (2015). Eye movement patterns in natural reading: A comparison of
monolingual and bilingual reading of a novel. PloS One, 10(8), e0134008.
Cowan, N. (2001). The magical number 4 in short-term memory: A reconsideration of mental storage
capacity. Behavioral and Brain Sciences, 24(1), 87-114.
Craik, F. 1. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal
of Verbal Learning and Verbal Behavior, 11, 671-684.
Cronan, T.H. (1987).* Reading rate: The measurement dilemma revisited. Forum for Reading, 19, 31-37.
Crossley, S. A., Skalicky, S., Dascalu, M., McNamara, D. S., & Kyle, K. (2017). Predicting text
comprehension, processing, and familiarity in adult readers: New approaches to readability formulas.
Discourse Processes, 54(5-6), 340-359.
Cupples, L., & Holmes, V. M. (1992).* Evidence for a difference in syntactic knowledge between skilled
and less skilled adult readers. Journal of Psycholinguistic Research, 21(4), 249-274.
Cushman, W. H. (1986).* Reading from microfiche, a VDT, and the printed page: subjective fatigue and
performance. Human factors, 28(1), 63-73.
Cutler, A., Norris, D., & Sebastián-Gallés, N. (2004). Phonemic repertoire and similarity within the
vocabulary. In 8th International Conference on Spoken Language Processing (Interspeech 2004-ICSLP)
(pp. 65-68). Sunjijn Printing Co.
Daniel, D.B., & Woody, W.D. (2013). E-textbooks at what cost? Performance and use of electronic v.
print texts. Computers & Education, 62(1), 18-23.
Deacon, S. H., Cook, K., & Parrila, R. (2012).* Identifying high-functioning dyslexics: is self-report of early
reading problems enough? Annals of Dyslexia, 62(2), 120-134.
Deal, R. W. (1934).* The development of reading and study habits in college students. Journal of
Educational Psychology, 25(4), 258-273.
Dee-Lucas, D. (1979).* Reading speed and memory for prose. Journal of Reading Behavior, 11(3), 221-
Delgado, P., Vargas, C., Ackerman, R., & Salmerón, L. (2018). Don't throw away your printed books: A
meta-analysis on the effects of reading media on reading comprehension. Educational Research Review,
Diependaele, K., Lemhöfer, K., & Brysbaert, M. (2013). The word frequency effect in first and second
language word recognition: A lexical entrenchment account. Quarterly Journal of Experimental
Dirix, N., Brysbaert, M., & Duyck, W. (2019).* How well do word recognition measures correlate? Effects
of language context and repeated presentations. Behavior Research Methods. Advance publication at
Dirix, N., Vander Beken, H., De Bruyne, E., Brysbaert, M., and Duyck, W. (2019). Reading text when
studying in a second language: An eye-tracking study. Manuscript submitted for publication.
Dixon, P., LeFevre, J. A., & Twilley, L. C. (1988).* Word knowledge and working memory as predictors of
reading skill. Journal of Educational Psychology, 80(4), 465-472.
Dumler, M. J. (1958).* A study of factors related to gains in the reading rate of college students trained
with the tachistoscope and accelerator. The Journal of Educational Research, 52(1), 27-30.
Dwyer, E. J., & West, R. F. (1994).* Effects of Sustained Silent Reading on Reading Rate among College
Students. Available at https://files.eric.ed.gov/fulltext/ED382924.pdf.
Dysli, M., Vogel, N., & Abegg, M. (2014).* Reading performance is not affected by a prism induced
increase of horizontal and vertical vergence demand. Frontiers in Human Neuroscience, 8, 431.
Dyson, M., & Haselgrove, M. (2000).* The effects of reading speed and reading patterns on the
understanding of text read from screen. Journal of Research in Reading, 23(2), 210-223.
Dyson, M. C., & Haselgrove, M. (2001).* The influence of reading speed and line length on the
effectiveness of reading from screen. International Journal of Human-Computer Studies, 54(4), 585-612.
Dyson, M. C., & Kipping, G. J. (1997).* The legibility of screen formats: are three columns better than
one? Computers & Graphics, 21(6), 703-712.
Elgort, I., Brysbaert, M., Stevens, M., & Van Assche, E. (2018). Contextual word learning during reading
in a second language: An eye-movement study. Studies in Second Language Acquisition, 40(2), 341-366.
Eurich, A. C., & Kraetsch, G. A. (1982). A 50-year comparison of University of Minnesota freshmen's
reading performance. Journal of Educational Psychology, 74(5), 660-665.
Everatt, J., & Underwood, G. (1994).* Individual differences in reading subprocesses: Relationships
between reading ability, lexical access, and eye movement control. Language and Speech, 37(3), 283-
Eviatar, Z., Ibrahim, R., Karelitz, T. M., & Simon, A. B. (2019).* Speed of reading texts in Arabic and
Hebrew. Reading and Writing, 32, 537-559.
Feng, S., D’Mello, S., & Graesser, A. C. (2013). Mind wandering while reading easy and difficult texts.
Psychonomic Bulletin & Review, 20(3), 586-592.
Fisher, D. F. (1975).* Reading and visual search. Memory & Cognition, 3(2), 188-196.
Ford, J. W., Missall, K. N., Hosp, J. L., & Kuhle, J. L. (2017). Examining oral passage reading rate across
three curriculum-based measurement tools for predicting grade-level proficiency. School Psychology
Review, 46(4), 363-378.
Folkmann Pedersen, H., Fusaroli, R., Lauridsen, L.L., & Parrila, R. (2016).* Reading Processes of University
Students with Dyslexia: An Examination of the Relationship between Oral Reading and Reading
Comprehension. Dyslexia, 22, 305–321
Foulke, E., & Sticht, T. G. (1969). Review of research on the intelligibility and comprehension of
accelerated speech. Psychological Bulletin, 72(1), 50-62.
Franken, G., Podlesek, A., & Možina, K. (2015).* Eye-tracking study of reading speed from LCD displays:
influence of type style and type size. Journal of Eye Movement Research, 8(1): 3, 1-8.
Fraser, C. A. (2007).* Reading rate in L1 Mandarin Chinese and L2 English across five reading tasks. The
Modern Language Journal, 91(3), 372-394.
Fredericks, T. K., Kumar, A. R., Oda, N. M., & Butt, S. E. (2015). Psychophysically determined user
acceptable oral reading speed (UAORS) for an 8-h work day. International Journal of Industrial
Ergonomics, 50, 52-61.
Freeburne, C. M., & Fleischer, M. S. (1952).* The effect of music distraction upon reading rate and
comprehension. Journal of Educational Psychology, 43(2), 101-109.
Frost, R. (1998). Toward a strong phonological theory of visual word recognition: true issues and false
trails. Psychological Bulletin, 123(1), 71-99.
Fry, E.B. (1963). Teaching faster reading: a manual. Cambridge: Cambridge University Press.
Fuchs, L. S., Fuchs, D., Hosp, M. K., & Jenkins, J. R. (2001). Oral reading fluency as an indicator of reading
competence: A theoretical, empirical, and historical analysis. Scientific Studies of Reading, 5(3), 239-256.
Fulford, C. P. (2001). A model of cognitive speed. International Journal of Instructional Media, 28(1), 31-
Galitz, W.O. (2007). The essential guide to user interface design: An introduction to GUI design principles
and techniques (3rd edition). Indianapolis, IN: Wiley Publishing Inc.
Ganayim, D., & Ibrahim, R. (2013).* How do typographical factors affect reading text and comprehension
performance in Arabic? Human Factors, 55(2), 323-332.
García, J. R., & Cain, K. (2014). Decoding and reading comprehension: A meta-analysis to identify which
reader and assessment characteristics influence the strength of the relationship in English. Review of
Educational Research, 84(1), 74-111.
Gibson, S. (2008). Reading aloud: a useful learning tool? ELT Journal, 62(1), 29-36.
Gilbert, L. C. (1959).* Speed of processing visual stimuli and its relation to reading. Journal of
Educational Psychology, 50(1), 8-14.
Glimne, S., Brautaset, R. L., & Seimyr, G. Ö. (2015).* The effect of glare on eye movements when
reading. Work, 50(2), 213-220.
Glock, M. D. (1949).* The effect upon eye-movements and reading rate at the college level of three
methods of training. Journal of Educational Psychology, 40(2), 93-106.
Goldman, S. R., & Saul, E. U. (1990). Flexibility in text processing: A strategy competition model. Learning
and Individual Differences, 2(2), 181-219.
Gordon-Salant, S., Zion, D. J., & Espy-Wilson, C. (2014). Recognition of time-compressed speech does not
predict recognition of natural fast-rate speech by older listeners. The Journal of the Acoustical Society of
America, 136(4), EL268-EL274.
Graf, P., & Levy, B. A. (1984).* Reading and remembering: Conceptual and perceptual processing
involved in reading rotated passages. Journal of Verbal Learning and Verbal Behavior, 23(3), 405-424.
Greene, E. B. (1931).* Effectiveness of various rates of silent reading of college students. Journal of
Applied Psychology, 15(2), 214-227.
Greene, E. B. (1934). Michigan speed of reading tests. The Journal of Educational Research, 28(4), 283-
Grisham, J. D., Sheppard, M. M., & Tran, W. U. (1993).* Visual symptoms and reading performance.
Optometry and Vision Science, 70(5), 384-391.
Gunraj, D. N., Drumm-Hewitt, A. M., & Klin, C. M. (2014).* Embodiment during reading: Simulating a
story character’s linguistic actions. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 40(2), 364-375.
Gunraj, D. N., & Klin, C. M. (2012).* Hearing story characters' voices: Auditory imagery during reading.
Discourse Processes, 49(2), 137-153.
Hage, D. S., & Stroud, J. B. (1959). Reading proficiency and intelligence scores, verbal and nonverbal. The
Journal of Educational Research, 52(7), 258-262.
Han, Q. M., Cong, L. J., Yu, C., & Liu, L. (2017).* Developing a Logarithmic Chinese Reading Acuity Chart.
Optometry and Vision Science, 94(6), 714-724.
Hanauer, D. (1998). The genre-specific hypothesis of reading: Reading poetry and encyclopedic items.
Poetics, 26(2), 63-80.
Harris, P. (2012).* Reading speed on paper vs computer vs Kindle. Vision Performance Institute. Available
Hartley, J. T. (1993).* Aging and prose memory: Tests of the resource-deficit hypothesis. Psychology and
Aging, 8(4), 538-551.
Hartley, J. T., Stojack, C. C., Mushaney, T. J., Annon, T. A., & Lee, D. W. (1994).* Reading speed and prose
memory in older and younger adults. Psychology and Aging, 9(2), 216-223.
Hasbrouck, J., & Tindal, G. A. (2006). Oral reading fluency norms: A valuable assessment tool for reading
teachers. The Reading Teacher, 59(7), 636-644.
Haught, P. A., & Walls, R. T. (2002).* Adult learners: New norms on the Nelson-Denny reading test for
healthcare professionals. Reading Psychology, 23(3), 217-238.
Hebert, M. E. (2017).* An Examination of Reading Comprehension and Reading Rate in University
Students. PhD thesis at the Department of Educational Psychology, University of Alberta. Available at
Hebert, M., Zhang, X., & Parrila, R. (2018).* Examining reading comprehension text and question
answering time differences in university students with and without a history of reading difficulties.
Annals of Dyslexia, 68(1), 15-24.
Heij, M. L. (2014).* The (in) effectiveness of pdf reading. Bachelor's thesis, University of Twente.
Available at http://essay.utwente.nl/65734/1/Heij,%20M.L.%20-%20s1189026%20(verslag).pdf.
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). Beyond WEIRD: Towards a broad-based behavioral
science. Behavioral and Brain Sciences, 33(2-3), 111-135.
Henry, L. K., & Lauer, A. R. (1939).* A comparison of four methods of increasing the reading speed of
college students. In Proceedings of the Iowa Academy of Science (Vol. 46, No. 1, pp. 273-276).
Henry, R., Van Dyke, J. A., & Kuperman, V. (2018).* Oculomotor planning in RAN and reading: a strong
test of the visual scanning hypothesis. Reading and Writing, 31(7), 1619-1643.
Hermena, E. W., Sheen, M., AlJassmi, M., AlFalasi, K., AlMatroushi, M., & Jordan, T. R. (2017).* Reading
rate and comprehension for text presented on tablet and paper: Evidence from Arabic. Frontiers in
Psychology, 8, 257.
Hess, T. M., & Tate, C. S. (1992).* Direct and indirect assessments of memory for script-based narratives
in young and older adults. Cognitive Development, 7(4), 467-484.
Hirai, A. (1999). The relationship between listening and reading rates of Japanese EFL learners. The
Modern Language Journal, 83(3), 367-384.
Hsieh, Y. C., Kuo, C. T., & Lin, H. (2016, July).* The Effect of Screen Size of Mobile Devices on Reading
Efficiency. In International Conference on Human Aspects of IT for the Aged Population (pp. 435-445).
Huang, D. L., Rau, P. L. P., & Liu, Y. (2009).* Effects of font size, display resolution and task type on
reading Chinese fonts from mobile devices. International Journal of Industrial Ergonomics, 39(1), 81-89.
Huey, E. B. (1901).* On the psychology and physiology of reading. II. The American Journal of
Psychology, 12(3), 292-312.
Huey, E.B. (1908). The Psychology and Pedagogy of Reading. Oxford, England: Macmillan.
Hunt, E. B., Davidson, J., & Lansman, M. (1981).* Individual differences in long-term memory access.
Memory & Cognition, 9(6), 599-608.
Hwang, S. L., Wang, M. Y., & Her, C. C. (1988).* An experimental study of Chinese information displays
on VDTs. Human Factors, 30(4), 461-471.
Hyönä, J., & Niemi, P. (1990).* Eye movements during repeated reading of a text. Acta Psychologica,
Hyönä, J., Lorch Jr, R. F., & Kaakinen, J. K. (2002).* Individual differences in reading to summarize
expository text: Evidence from eye fixation patterns. Journal of Educational Psychology, 94(1), 44-55.
İdil, Ş. A., Çaliskan, D., & İdil, N. B. (2011).* Development and validation of the Turkish version of the
MNREAD visual acuity charts. Turkish Journal of Medical Sciences, 41(4), 565-570.
Jackson, J. D., & Balota, D. A. (2012).* Mind-wandering in younger and older adults: Converging
evidence from the sustained attention to response task and reading for comprehension. Psychology and
Aging, 27(1), 106.
Jackson, M. D. (1980). Further evidence for a relationship between memory access and reading ability.
Journal of Verbal Learning and Verbal Behavior, 19(6), 683-694.
Jackson, M. D., & McClelland, J. L. (1975). Sensory and cognitive determinants of reading speed. Journal
of Verbal Learning and Verbal Behavior, 14(6), 565-574.
Jackson, M. D., & McClelland, J. L. (1979). Processing determinants of reading speed. Journal of
Experimental Psychology: General, 108(2), 151-181.
Jackson, N. E., Chen, H., Goldsberry, L., Kim, A., & Vanderwerff, C. (1999).* Effects of variations in
orthographic information on Asian and American readers' English text reading. Reading and Writing,
Jensen, P. K., Mills, J. W., & Hershkowitz, M. (1971).* An Evaluation of Four Methods of Reading
Instruction as a Function of Time of Testing. Journal of Reading Behavior, 4(3), 32-37.
Jester, R. E., & Travers, R. M. (1966). Comprehension of connected meaningful discourse as a function of
rate and mode of presentation. The Journal of Educational Research, 59(7), 297-302.
Johansson, J., Pansell, T., Ygge, J., & Seimyr, G. Ö. (2014).* Monocular and binocular reading
performance in subjects with normal binocular vision. Clinical and Experimental Optometry, 97(4), 341-
Johns, B.T., & Dye, M. (2019). Gender bias at scale: Evidence from the usage of personal nouns. Behavior
Johnson, R. L., Bui, B., & Schmitt, L. L. (2018).* Are two spaces better than one? The effect of spacing
following periods and commas during reading. Attention, Perception, & Psychophysics, 80(6), 1504-1511.
Jonassen, D.H. (Ed.) (2004). Handbook of research on educational communications and technology (2nd
edition). Mahwah, NJ: Lawrence Erlbaum Associates.
Jones, J. S., Conradi, K., & Amendum, S. J. (2016). Matching interventions to reading needs: A case for
differentiation. The Reading Teacher, 70(3), 307-316.
Jones, L. Ø., Varberg, J., Manger, T., Eikeland, O. J., & Asbjørnsen, A. (2012).* Reading and writing self-
efficacy of incarcerated adults. Learning and Individual Differences, 22(3), 343-349.
Jordan, T. R., Almabruk, A. A., Gadalla, E. A., McGowan, V. A., White, S. J., Abedipour, L., & Paterson, K.
B. (2014). Reading direction and the central perceptual span: Evidence from Arabic and English.
Psychonomic bulletin & review, 21(2), 505-511.
Jufri, S., Buari, N. H., & Chen, A. H. (2016).* Text structures affect reading speed. Social and
Management Research Journal, 13(1), 118-131.
Juola, J. F., Ward, N. J., & McNamara, T. (1982).* Visual search and reading of rapid serial presentations
of letter strings, words, and text. Journal of Experimental Psychology: General, 111(2), 208-227.
Just, M.A., & Carpenter, P.A. (1987).* The psychology of reading and language comprehension. Boston:
Allyn and Bacon, Inc.
Kamienkowski, J. E., Carbajal, M. J., Bianchi, B., Sigman, M., & Shalom, D. E. (2018).* Cumulative
Repetition Effects across Multiple Readings of a Word: Evidence From Eye Movements. Discourse
Processes, 55(3), 256-271.
Kang, Y. Y., Wang, M. J. J., & Lin, R. (2009).* Usability evaluation of e-books. Displays, 30(2), 49-52.
Karakus, S., Mathews, P. M., Agrawal, D., Henrich, C., Ramulu, P. Y., & Akpek, E. K. (2018).* Impact of Dry
Eye on Prolonged Reading. Optometry and Vision Science, 95(12), 1105-1113.
Kasperski, R., Shany, M., & Katzir, T. (2016). The role of RAN and reading rate in predicting reading self-
concept. Reading and Writing, 29(1), 117-136.
Kass, R E., & Wasserman, L. (1995). A reference Bayesian test for nested hypotheses and its relationship
to the Schwarz criterion. Journal of the American Statistical Association, 90, 928-934.
Kemper, S., Jackson, J. D., Cheung, H., & Anagnopoulos, C. A. (1993). Enhancing older adults' reading
comprehension. Discourse processes, 16(4), 405-428.
Kemper, S., & Sumner, A. (2001). The structure of verbal abilities in young and older adults. Psychology
and Aging, 16(2), 312-322.
King, P. E., & Behnke, R. R. (1989). The effect of time-compressed speech on comprehensive,
interpretive, and short-term listening. Human Communication Research, 15(3), 428-443.
King, P. T., Dellande, W. D., & Walter, T. L. (1969).* The prediction of change in grade point average from
initial reading rates. Journal of Reading, 13(3), 215-246.
Kingston, A. J., & George, C. E. (1957).* The Effects of Special Reading Training Upon the Development of
College Students’ Reading Skills. The Journal of Educational Research, 50(6), 471-475.
Kintsch, W., & Monk, D. (1972).* Storage of complex information in memory: Some implications of the
speed with which inferences can be made. Journal of Experimental Psychology, 94(1), 25-32.
Kintsch, W., & van Dijk, T. A. (1978). Toward a model of text comprehension and production.
Psychological Review, 85, 363-394.
Kirby, J. R., Georgiou, G. K., Martinussen, R., & Parrila, R. (2010). Naming speed and reading: From
prediction to instruction. Reading Research Quarterly, 45(3), 341-362.
Kirby, J. R., Silvestri, R., Allingham, B. H., Parrila, R., & La Fave, C. B. (2008).* Learning strategies and
study approaches of postsecondary students with dyslexia. Journal of Learning Disabilities, 41(1), 85-96.
Kliegl, R., Grabner, E., Rolfs, M., & Engbert, R. (2004). Length, frequency, and predictability effects of
words on eye movements in reading. European Journal of Cognitive Psychology, 16(1-2), 262-284.
Kong, Y., Seo, Y. S., & Zhai, L. (2018). Comparison of reading performance on screen and on paper: A
meta-analysis. Computers & Education, 123, 138-149.
Korinth, S. P., & Fiebach, C. J. (2018).* Improving Silent Reading Performance Through Feedback on Eye
Movements: A Feasibility Study. Scientific Studies of Reading, 22(4), 289-307.
Kruk, R. S., & Muter, P. (1984).* Reading of continuous text on video screens. Human Factors, 26(3),
Kuperman, V., Siegelman, N., & Frost, R. (2019).* Information distribution and reading time across
writing systems. Manuscript submitted for publication.
Laasonen, M., Service, E., & Virsu, V. J. (2001).* Temporal order and processing acuity of visual, auditory,
and tactile perception in developmentally dyslexic young adults. Cognitive, Affective, & Behavioral
Neuroscience, 1(4), 394-410.
Lajoie, D. L. (2013).* Reading Comprehension and Nursing Education: A Missing Variable Associated with
Nursing Student Attrition? PhD Thesis University of Wisconsin Milwaukee. Available at
Laubrock, J., & Kliegl, R. (2015).* The eye-voice span during reading aloud. Frontiers in Psychology, 6,
Lee, Y.S. (2003).* A study of reading with a handheld computer. PhD Thesis Virginia Polytechnic Institute
and State University. Available at
Leinonen, S., Müller, K., Leppänen, P. H., Aro, M., Ahonen, T., & Lyytinen, H. (2001).* Heterogeneity in
adult dyslexic readers: Relating processing skills to the speed and accuracy of oral text reading. Reading
and Writing, 14(3-4), 265-296.
Letson, C. T. (1958).* Speed and comprehension in reading. The Journal of Educational Research, 52(2),
Lewandowski, L. J., Codding, R. S., Kleinmann, A. E., & Tucker, K. L. (2003).* Assessment of reading rate
in postsecondary students. Journal of Psychoeducational Assessment, 21(2), 134-144.
Lewandowsky, S., & Oberauer, K. (2015). Rehearsal in serial recall: An unworkable solution to the
nonexistent problem of decay. Psychological Review, 122(4), 674-699.
R. J. Light; D. B. Pillemer (1984). Summing up: The Science of Reviewing Research. Cambridge,
Massachusetts.: Harvard University Press.
Lijka, B., Toor, S., & Arblaster, G. (2019).* The Impact of Diplopia on Reading. British and Irish Orthoptic
Journal, 15(1), 8-14.
Lindgren, S. A., & Laine, M. (2011).* Multilingual dyslexia in university students: Reading and writing
patterns in three languages. Clinical Linguistics & Phonetics, 25(9), 753-766.
Liversedge, S. P., Drieghe, D., Li, X., Yan, G., Bai, X., & Hyönä, J. (2016).* Universality in eye movements
and reading: A trilingual investigation. Cognition, 147, 1-20.
Lloyd, J., & McKelvie, S. J. (1992).* Effects of spatial disruptions on reading speed for fast and slow
readers. The Journal of General Psychology, 119(3), 229-235.
Lott, L. A., Schneck, M. E., Haegerström-Portnoy, G., Brabyn, J. A., Gildengorin, G. L., & West, C. G.
(2001). Reading performance in older adults with good acuity. Optometry and Vision Science, 78(5), 316-
Lutz, K. C., & Mallard, A. R. (1986).* Disfluencies and rate of speech in young adult nonstutterers. Journal
of Fluency Disorders, 11(4), 307-316.
Mackensen, G., & Stichler, H. (1963).* Untersuchungen der Lesegeschwindigkeit in Abhängigkeit vom
Bildungsgrad. Albrecht von Graefes Archiv für Ophthalmologie, 166(1), 81-86.
Mak, M., & Willems, R. M. (2018).* Mental simulation during literary reading: Individual differences
revealed with eye-tracking. Language, Cognition and Neuroscience, 34(4), 511-535.
Maki, R. H., Jonas, D., & Kallod, M. (1994).* The relationship between comprehension and
metacomprehension ability. Psychonomic Bulletin & Review, 1(1), 126-129.
Marian, V., Bartolotti, J., Chabal, S., & Shook, A. (2012). CLEARPOND: Cross-linguistic easy-access
resource for phonological and orthographic neighborhood densities. PloS One, 7(8), e43230.
Martin-Chang, S. L., & Gould, O. N. (2008).* Revisiting print exposure: Exploring differential links to
vocabulary, comprehension and reading rate. Journal of Research in Reading, 31(3), 273-284.
Marx, T. (2015).* Vergleich zweier deutschsprachiger, standardisierter Lesetafeln zur Bestimmung der
Lesegeschwindigkeit. PhD Thesis. University of Tuebingen. Available at https://publikationen.uni-
Masson, M. E. (1982).* Cognitive processes in skimming stories. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 8(5), 400-417.
Masterson, J., & Hayes, M. (2004).* UK data from 197 undergraduates for the Nelson Denny Reading
Test. Journal of Research in Reading, 27(1), 30-35.
Mataftsi, A., Bourtoulamaiou, A., Haidich, A. B., Antoniadis, A., Kilintzis, V., Tsinopoulos, I. T., &
Dimitrakos, S. (2013).* Development and validation of the Greek version of the MNREAD acuity chart.
Clinical and Experimental Optometry, 96(1), 25-31.
Mathews, P. M., Ramulu, P. Y., Swenor, B. S., Utine, C. A., Rubin, G. S., & Akpek, E. K. (2017).* Functional
impairment of reading in patients with dry eye. British Journal of Ophthalmology, 101(4), 481-486.
Maxwell, M. J., & Mueller, A. C. (1965).* An experiment on the effect of motivational appeal vs.
techniques upon reading rate improvement in a group of college students. Counseling Center University
of Mariland. Available at https://files.eric.ed.gov/fulltext/ED011490.pdf.
Mayr, S., Köpper, M., & Buchner, A. (2017).* Effects of high pixel density on reading comprehension,
proofreading performance, mood state, and physical discomfort. Displays, 48, 41-49.
McCollum, J. C. (2012).* The Correlation of Arab ELLs' Academic Reading Fluency in Arabic and English.
Master Thesis, Brigham Young University. Available at: https://scholarsarchive.byu.edu/etd/3657/.
McConkie, G. W., & Meyer, B. J. (1974).* Investigation of reading strategies: II. A replication of payoff
condition effects. Journal of Reading Behavior, 6(2), 151-158.
McConkie, G. W., & Rayner, K. (1974).* Investigation of reading strategies: I. Manipulating strategies
through payoff conditions. Journal of Reading Behavior, 6(1), 9-18.
McConkie, G. W., Rayner, K., & Wilson, S. J. (1973).* Experimental manipulation of reading strategies.
Journal of Educational Psychology, 65(1), 1-8.
Mellard, D., Woods, K., & Md Desa, Z. (2012).* An oral reading fluency assessment for young adult
career and technical education students. Learning Disabilities Research & Practice, 27(3), 125-135.
Meyer, B. J., Talbot, A. P., & Florencio, D. (1999). Reading rate and prose retrieval. Scientific Studies of
Reading, 3(4), 303-329.
Miller, G. R., & Coleman, E. B. (1971).* The measurement of reading speed and the obligation to
generalize to a population of reading materials. Journal of Reading Behavior, 4(3), 48-56.
Miller-Guron, L., & Lundberg, I. (2000).* Dyslexia and second language reading: A second bite at the
apple? Reading and Writing, 12(1-2), 41-61.
Miranda, A. M., Nunes-Pereira, E. J., Baskaran, K., & Macedo, A. F. (2018).* Eye movements,
convergence distance and pupil-size when reading from smartphone, computer, print and tablet.
Scandinavian Journal of Optometry and Visual Science, 11(1), 1-5.
Mokhtari, K., & Reichard, C. A. (2002). Assessing students' metacognitive awareness of reading
strategies. Journal of Educational Psychology, 94(2), 249-259.
Monk, A. F. (1984).* Reading continuous text from a one-line visual display. International Journal of
Man-Machine Studies, 21(3), 269-277.
Monteiro de Castro, C. T., Kallie, C. S., & Salomão, S. R. (2005).* Development and validation of the
MNREAD reading acuity chart in Portuguese. Arquivos Brasileiros de Oftalmologia, 68(6), 777-783.
Morrice, E. (2017).* Validation of the International Reading Speed Texts in a Canadian Sample. Master
Thesis, Concordia University. Available at
Moys, J. L., Loveland, P., & Dyson, M. C. (2018).* eInk versus paper: exploring the effects of medium and
typographic quality on recall and reading speed. Visible Language, 52(3), 75-95.
Munch, I. C., Jørgensen, A. H. R., & Radner, W. (2016).* The Danish version of the R adner R eading C
hart: design and empirical testing of sentence optotypes in subjects of varying educational background.
Acta Ophthalmologica, 94(2), 182-186.
Muter, P., Latrémouille, S. A., Treurniet, W. C., & Beam, P. (1982).* Extended reading of continuous text
on television screens. Human Factors, 24(5), 501-508.
Muter, P., & Maurutto, P. (1991).* Reading and skimming from computer screens and books: the
paperless office revisited? Behaviour & Information Technology, 10(4), 257-266.
Na, N., Choi, K., & Suk, H. J. (2016).* Adaptive luminance difference between text and background for
comfortable reading on a smartphone. International Journal of Industrial Ergonomics, 51, 68-72.
Naveh-Benjamin, M., & Ayres, T. J. (1986). Digit span, reading rate, and linguistic relativity. The Quarterly
Journal of Experimental Psychology Section A, 38(4), 739-751.
Nederlandse Nieuwsmonitor (2014). Minder nieuws voor hetzelfde geld? Van broadsheet naar tabloid.
Available at http://nieuwsmonitor.org/wp-content/uploads/Tabloid_juni.pdf.
Nelson, M., & Denny, E. (1929). The Nelson-Denny Reading Test for colleges and senior high schools.
New York: Houghton Mifflin.
Nergård-Nilssen, T., & Hulme, C. (2014).* Developmental dyslexia in adults: Behavioural manifestations
and cognitive correlates. Dyslexia, 20(3), 191-207.
Niple, M. L. D. (1968).* The relationship of different study methods to immediate and delayed
comprehension. Doctoral dissertation, Ohio State University. Available at
Noyes, J. M., & Garland, K. J. (2003).* VDT versus paper-based text: Reply to Mayes, Sims and Koonce.
International Journal of Industrial Ergonomics, 31(6), 411-423.
Oliver, W. L., Healy, A. F., & Mross, E. F. (2005).* Trade-offs in detecting letters and comprehending text.
Canadian Journal of Experimental Psychology/Revue Canadienne de Psychologie Expérimentale, 59(3),
Öquist, G., & Goldstein, M. (2003).* Towards an improved readability on mobile devices: evaluating
adaptive rapid serial visual presentation. Interacting with Computers, 15(4), 539-558.
Otten, M. (2015).* Evaluation of the workshop Effectief studeren and the online course Snellezen. Master
Thesis University of Twente. Available at https://docplayer.nl/19192070-Evaluation-of-the-workshop-
Ousler III, G. W., Rodriguez, J. D., Smith, L. M., Lane, K. J., Heckley, C., Angjeli, E., & Abelson, M. B.
(2015).* Optimizing reading tests for dry eye disease. Cornea, 34(8), 917-921.
Owsley, C. (2011). Aging and vision. Vision Research, 51, 1610–1622.
Palmer, J., MacLeod, C. M., Hunt, E., & Davidson, J. E. (1985). Information processing correlates of
reading. Journal of Memory and Language, 24(1), 59-88.
Pan, J., Laubrock, J., & Yan, M. (2016).* Parafoveal processing in silent and oral reading: Reading mode
influences the relative weighting of phonological and semantic information in Chinese. Journal of
Experimental Psychology: Learning Memory and Cognition, 42(8), 1257–1273.
Pan, J., Yan, M., & Laubrock, J. (2017).* Perceptual span in oral reading: The case of Chinese. Scientific
Studies of Reading, 21(3), 254-263.
Pan, J., Yan, M., Laubrock, J., & Shu, H. (2019).* Lexical and sublexical phonological effects in chinese
silent and oral reading. Scientific Studies of Reading. Advance publication available at DOI:
Park, Y. (2016).* Technology-Enhanced L2 Reading: The Effects of Hierarchical Phrase Segmentation. PhD
Thesis at University of California, Irvine. Available at https://escholarship.org/uc/item/9st803r1.
Pashler, H., Kang, S. H., & Ip, R. Y. (2013).* Does multitasking impair studying? Depends on timing.
Applied Cognitive Psychology, 27(5), 593-599.
Paterson, K. B., Almabruk, A. A., McGowan, V. A., White, S. J., & Jordan, T. R. (2015).* Effects of word
length on eye movement control: The evidence from Arabic. Psychonomic Bulletin & Review, 22(5),
Paterson, K. B., & Jordan, T. R. (2010).* Effects of increased letter spacing on word identification and eye
guidance during reading. Memory & Cognition, 38(4), 502-512.
Paterson, K. B., McGowan, V. A., White, S. J., Malik, S., Abedipour, L., & Jordan, T. R. (2014).* Reading
direction and the central perceptual span in Urdu and English. PloS One, 9(2), e88358.
Perea, M., & Acha, J. (2009).* Space information is important for reading. Vision Research, 49(15), 1994-
Perfetti, C. A. (1985). Reading ability. New York: Oxford Press.
Perrin, J. L., Paillé, D., & Baccino, T. (2014).* Reading tilted: Does the use of tablets impact performance?
An oculometric study. Computers in Human Behavior, 39, 339-345.
Perrin, J. L., Paillé, D., & Baccino, T. (2015). A new sentence generator providing material for maximum
reading speed measurement. Behavior Research Methods, 47(4), 1055-1064.
Persky, A. M., & Hogg, A. (2017). Influence of reading material characteristics on study time for pre-class
quizzes in a flipped classroom. American Journal of Pharmaceutical Education, 81(6), Article 103.
Pollatsek, A., Bolozky, S., Well, A. D., & Rayner, K. (1981).* Asymmetries in the perceptual span for Israeli
readers. Brain and Language, 14(1), 174-180.
Potter, C. A. (1954).* Evaluation of the Reading Program of the US Naval Academy During Summer 1951.
Educational and Psychological Measurement, 14(1), 193-203.
Poulton, E. C. (1961).* British courses for adults on effective reading. British Journal of Educational
Psychology, 31, 128-137.
Preston, R. C., & Botel, M. (1952).* The relation of reading skill and other factors to the academic
achievement of 2048 college students. The Journal of Experimental Education, 20(4), 363-371.
Preston, R. C., & Tuft, E. N. (1948).* The reading habits of superior college students. The Journal of
Experimental Education, 16(3), 196-202.
Primativo, S., Spinelli, D., Zoccolotti, P., De Luca, M., & Martelli, M. (2016). Perceptual and cognitive
factors imposing “speed limits” on reading rate: a study with the rapid serial visual presentation. PloS
One, 11(4), e0153786.
Quantz, J. O. (1898). Problems in the psychology of reading. Psychological Review, 5(4), 434-436.
Radner, W., & Diendorfer, G. (2014).* English sentence optotypes for measuring reading acuity and
speed—the English version of the RADNER Reading Charts. Graefe's Archive for Clinical and Experimental
Ophthalmology, 252(8), 1297-1303.
Radner, W., Obermayer, W., Richter-Mueksch, S., Willinger, U., Velikay-Parel, M., & Eisenwort, B. (2002).
The validity and reliability of short German sentences for measuring reading speed. Graefe's archive for
clinical and experimental ophthalmology, 240(6), 461-467.
Radner, W., Radner, S., & Diendorfer, G. (2016).* A new principle for the standardization of long
paragraphs for reading speed analysis. Graefe's Archive for Clinical and Experimental Ophthalmology,
Ramadan, M. Z. (2011).$ Evaluating college students’ performance of Arabic typeface style, font size,
page layout and foreground/background color combinations of e-book materials. Journal of King Saud
University-Engineering Sciences, 23(2), 89-100.
Ramulu, P. Y., Swenor, B. K., Jefferys, J. L., & Rubin, G. S. (2013).* Description and validation of a test to
evaluate sustained silent reading. Investigative Ophthalmology & Visual Science, 54(1), 673-680.
Rasinski, T. V., & Hoffman, J. V. (2003). Oral reading in the school literacy curriculum. Reading Research
Quarterly, 38(4), 510-522.
Rastle, K., & Brysbaert, M. (2006). Masked phonological priming effects in English: Are they real? Do
they matter? Cognitive Psychology, 53(2), 97-145.
Rayner, K. (1975). The perceptual span and peripheral cues in reading. Cognitive Psychology, 7(1), 65-81.
Rayner, K. (1978). Eye movements in reading and information processing. Psychological Bulletin, 85(3),
Rayner, K. (1986).* Eye movements and the perceptual span in beginning and skilled readers. Journal of
Experimental Child Psychology, 41(2), 211-236.
Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research.
Psychological Bulletin, 124(3), 372-422.
Rayner, K., Fischer, M. H., & Pollatsek, A. (1998).* Unspaced text interferes with both word identification
and eye movement control. Vision Research, 38(8), 1129-1144.
Rayner, K., & Pollatsek, A. (1989). The Psychology of Reading. Englewood Cliffs, NJ Prentice Hall.
Rayner, K., Pollatsek, A., Ashby, J., & Clifton Jr, C. (2012). Psychology of reading (2
edition). London UK:
Rayner, K., Reichle, E. D., Stroud, M. J., Williams, C. C., & Pollatsek, A. (2006). The effect of word
frequency, word predictability, and font difficulty on the eye movements of young and older readers.
Psychology and Aging, 21, 448–465.
Rayner, K., Schotter, E. R., Masson, M. E., Potter, M. C., & Treiman, R. (2016). So much to read, so little
time: How do we read, and can speed reading help? Psychological Science in the Public Interest, 17(1), 4-
Rayner, K., Slattery, T. J., & Bélanger, N. N. (2010).* Eye movements, the perceptual span, and reading
speed. Psychonomic Bulletin & Review, 17(6), 834-839.
Re, A. M., Tressoldi, P. E., Cornoldi, C., & Lucangeli, D. (2011).* Which tasks best discriminate between
dyslexic university students and controls in a transparent language? Dyslexia, 17(3), 227-241.
Reichle, E. D., Liversedge, S. P., Drieghe, D., Blythe, H. I., Joseph, H. S., White, S. J., & Rayner, K. (2013).
Using EZ Reader to examine the concurrent development of eye-movement control and reading skill.
Developmental Review, 33(2), 110-149.
Reichle, E. D., Pollatsek, A., & Rayner, K. (2012). Using EZ Reader to simulate eye movements in
nonreading tasks: A unified framework for understanding the eye–mind link. Psychological Review,
Rello, L., & Baeza-Yates, R. (2016).* The effect of font type on screen readability by people with dyslexia.
ACM Transactions on Accessible Computing (TACCESS), 8(4), 15.
Reuter-Lorenz, P. A., & Cappell, K. A. (2008). Neurocognitive aging and the compensation hypothesis.
Current Directions in Psychological Science, 17(3), 177-182.
Revelle, W., & Condon, D. M. (2018). Reliability from alpha to omega: a tutorial. Available at
Rhiu, S., Kim, M., Kim, J. H., Lee, H. J., & Lim, T. H. (2017).* Korean Version Self-testing Application for
Reading Speed. Korean Journal of Ophthalmology, 31(3), 202-208.
Ricciardi, O., & Di Nocera, F. (2017).* Not so fast: A reply to Benedetto et al.(2015). Computers in Human
Behavior, 69, 381-385.
Roberts, E. O., Proudlock, F. A., Martin, K., Reveley, M. A., Al-Uzri, M., & Gottlob, I. (2012).* Reading in
schizophrenic subjects and their nonsymptomatic first-degree relatives. Schizophrenia Bulletin, 39(4),
Rodero, E. (2012). A comparative analysis of speech rate and perception in radio bulletins. Text & Talk,
Rodero, E. (2016). Influence of speech rate and information density on recognition: The moderate
dynamic mechanism. Media Psychology, 19(2), 224-242.
Rothkopf, E. Z., & Billington, M. J. (1983). Passage length and recall with test size held constant: Effects
of modality, pacing, and learning set. Journal of Verbal Learning and Verbal Behavior, 22(6), 667-681.
Rosa, A. M., Farinha, C. L., Radner, W., Diendorfer, G., Loureiro, M. F., & Murta, J. N. (2016).*
Development of the Portuguese version of a standardized reading test: the Radner-Coimbra Charts.
Arquivos Brasileiros de Oftalmologia, 79(4), 238-242.
Rose, F. C., & Rostas, S. M. (1946).* The effect of illumination on reading rate and comprehension of
college students. Journal of Educational Psychology, 37(5), 279-292.
Rouder, J. N., & Haaf, J. M. (2018). Power, dominance, and constraint: A note on the appeal of different
design traditions. Advances in Methods and Practices in Psychological Science, 1(1), 19-26.
Rubin, G. S., & Turano, K. (1992). Reading without saccadic eye movements. Vision Research, 32(5), 895-
Ruppel, G. (1979).* Self-management and reading rate improvement. Journal of Counseling Psychology,
Sackstein, S., Spark, L., & Jenkins, A. (2015).* Are e-books effective tools for learning? Reading speed and
comprehension: iPad® vs. paper. South African Journal of Education, 35(4), 1-14..
Sainio, M., Hyönä, J., Bingushi, K., & Bertram, R. (2007).* The role of interword spacing in reading
Japanese: An eye movement study. Vision Research, 47(20), 2575-2584.
Salthouse, T. A. (1993). Speed and knowledge as determinants of adult age differences in verbal tasks.
Journal of Gerontology, 48(1), P29-P36.
Samuels, S. J., & Dahl, P. R. (1975).* Establishing Appropriate Purpose for Reading and Its Effect on
Flexibility of Reading Rate. Journal of Educational Psychology, 67(1), 38-43.
Savage, R., & Frederickson, N. (2005). Evidence of a highly specific relationship between rapid automatic
naming of digits and text-reading speed. Brain and Language, 93(2), 152-159.
Scrucca, L., Fop, M., Murphy, T. B., & Raftery, A. E. (2016). Mclust 5: clustering, classification and density
estimation using Gaussian finite mixture models. The R journal, 8(1), 289.
Seidenberg, M. (2017). Language at the speed of sight: How we read, why so many can’t, and what can
be done about it. Hachette Book Group, Inc: Basic Books.
Sekey, A., & Tietz, J. (1982).* Text display by" saccadic scrolling". Visible Language, 16(1), 62-77.
Seo, H., & Lee, C. (2002).* Head-free reading of horizontally and vertically arranged texts. Vision
Research, 42(10), 1325-1337.
Shankweiler, D., & Fowler, C. A. (2019). Relations Between Reading and Speech Manifest Universal
Phonological Principle. Annual Review of Linguistics, 5, 109-129.
Share, D. L. (2008). On the Anglocentricities of current reading research and practice: the perils of
overreliance on an" outlier" orthography. Psychological Bulletin, 134(4), 584-615.
Sheedy, J. E., Subbaram, M. V., & Hayes, J. R. (2003).* Filters on computer displays–effects on legibility,
performance and comfort. Behaviour and Information Technology, 22(6), 427-433.
Sheedy, J., Tai, Y. C., Subbaram, M., Gowrisankaran, S., & Hayes, J. (2008).* ClearType sub-pixel text
rendering: Preference, legibility and reading performance. Displays, 29(2), 138-151.
Shimoda, T. A. (1993).* The effects of interesting examples and topic familiarity on text comprehension,
attention, and reading speed. The Journal of Experimental Education, 61(2), 93-103.
Shimron, J., & Sivan, T. (1994).* Reading proficiency and orthography evidence from Hebrew and
English. Language Learning, 44(1), 5-27.
Shu, H., Zhou, W., Yan, M., & Kliegl, R. (2011).* Font size modulates saccade-target selection in Chinese
reading. Attention, Perception, & Psychophysics, 73(2), 482-490.
Siegenthaler, E., Bochud, Y., Bergamin, P., & Wurtz, P. (2012).* Reading on LCD vs e-Ink displays: effects
on fatigue and visual strain. Ophthalmic and Physiological Optics, 32(5), 367-374.
Siegenthaler, E., Wurtz, P., Bergamin, P., & Groner, R. (2011).* Comparing reading processes on e-ink
displays and print. Displays, 32(5), 268-273.
Simola, J., Salojärvi, J., & Kojo, I. (2008). Using hidden Markov model to uncover processing states from
eye movements in information search tasks. Cognitive Systems Research, 9(4), 237-251.
Singer Trakhman, L. M., Alexander, P. A., & Berkowitz, L. E. (2017).* Effects of processing time on
comprehension and calibration in print and digital mediums. The Journal of Experimental Education.
Advance Publication available at https://doi.org/10.1080/00220973.2017.1411877.
Singer Trakhman, L. M., Alexander, P. A., & Silverman, A. B. (2018). Profiling reading in print and digital
mediums. Learning and Instruction, 57, 5-17.
Slattery, T. J., & Rayner, K. (2010).* The influence of text legibility on eye movements during reading.
Applied Cognitive Psychology, 24(8), 1129-1148.
Slattery, T. J., & Yates, M. (2017).* Word skipping: effects of word length, predictability, spelling and
reading skill. The Quarterly Journal of Experimental Psychology. Advance Publication available at
Smith, D. E., & Wood, R. L. (1955).* Reading improvement and college grades: a follow-up. Journal of
Educational Psychology, 46(3), 151-159.
Smith, M.C., & Pourchot, T. (Eds.) (1998). Adult learning and development: Perspectives from
educational psychology. Mahwah, N.J: L. Erlbaum.
Sobell, L. C., & Sobell, M. B. (1972).* Effects of alcohol on the speech of alcoholics. Journal of Speech and
Hearing Research, 15(4), 861-868.
Sobell, L. C., Sobell, M. B., & Colemanb, R. F. (1982).* Alcohol-Induced Dysﬂuency in Nonalcoholics. Folia
Phoniatrica, 34, 316-323.
Song, J., Kim, J. H., & Hyung, S. (2016).* Validity of Korean version reading speed application and
measurement of reading speed: pilot study. Journal of the Korean Ophthalmological Society, 57(4), 642-
Spache, G. D. (1962). Is this a breakthrough in reading? The Reading Teacher, 15(4), 258-263.
Sperling, G. (1960). The information available in brief visual presentations. Psychological Monographs:
General and Applied, 74(11), 1-29.
Spichtig, A. N., Hiebert, E. H., Vorstius, C., Pascoe, J. P., David Pearson, P., & Radach, R. (2016). The
Decline of Comprehension-Based Silent Reading Efficiency in the United States: A Comparison of Current
Data With Performance in 1960. Reading Research Quarterly, 51(2), 239-259.
Spragins, A. B., Lefton, L. A., & Fisher, D. F. (1976).* Eye movements while reading and searching
spatially transformed text: A developmental examination. Memory & Cognition, 4(1), 36-42.
Sprenger-Charolles, L., Siegel, L. S., Béchennec, D., & Serniclaes, W. (2003). Development of phonological
and orthographic processing in reading aloud, in silent reading, and in spelling: A four-year longitudinal
study. Journal of Experimental Child Psychology, 84(3), 194-217.
Standing, L., & Curtis, L. (1989).* Subvocalization rate versus other predictors of the memory span.
Psychological Reports, 65(2), 487-495.
Stanley, T. D., Carter, E. C., & Doucouliagos, H. (2018). What meta-analyses reveal about the replicability
of psychological research. Psychological Bulletin, 144(12), 1325-1346.
Stine-Morrow, E. A., Miller, L. M. S., & Hertzog, C. (2006). Aging and self-regulated language processing.
Psychological Bulletin, 132(4), 582-606.
Stoll, F. (1974).* Évaluation de trois types d'exercices de lecture rapide. Le Travail Humain, 37(2), 249-
Stroud, J. B. (1945). Rate of visual perception as a factor in rate of reading. Journal of Educational
Psychology, 36(8), 487-498.
Strukelj, A., & Niehorster, D. C. (2018).* One page of text: Eye movements during regular and thorough
reading, skimming, and spell checking. Journal of Eye Movement Research, 11(1):1-22.
Suárez-Coalla, P., & Cuetos, F. (2015).* Reading difficulties in Spanish adults with dyslexia. Annals of
Dyslexia, 65(1), 33-51.
Subbaram, V. M. (2004).* Effect of display and text parameters on reading performance. Doctoral
dissertation, Ohio State University. Available at
Subramanian, A., & Pardhan, S. (2006).* The repeatability of MNREAD acuity charts and variability at
different test distances. Optometry and Vision Science, 83(8), 572-576.
Sun, F., Morita, M., & Stark, L. W. (1985).* Comparative patterns of reading eye movement in Chinese
and English. Perception & Psychophysics, 37(6), 502-506.
Tauroza, S., & Allison, D. (1990). Speech rates in British English. Applied Linguistics, 11(1), 90-105.
Taylor, E. A. (1957). The Spans: Perception, Apprehension, and Recognition As Related to Reading and
Speed Reading. American Journal of Ophthalmology, 44(4), 501-507.
Taylor, S. E. (1965).* Eye movements in reading: Facts and fallacies. American Educational Research
Journal, 2(4), 187-202.
Thalberg, S. P. (1967).* Reading rate and immediate versus delayed retention. Journal of Educational
Psychology, 58(6), 373-378.
Thames, K. H., & Rossiter, C. M. (1972).* The effects of reading practice with compressed speech on
reading rate and listening comprehension. AV Communication Review, 20(1), 35-42.
Tinker, M. A. (1939). Speed versus comprehension in reading as affected by level of difficulty. Journal of
Educational Psychology, 30(2), 81-94.
Tinker, M. A. (1955). Prolonged reading tasks in visual research. Journal of Applied Psychology, 39(6),
Tinker, M. A. (1958). Recent studies of eye movements in reading. Psychological Bulletin, 55(4), 215-231.
Tombaugh, J. W., Arkin, M. D., & Dillion, R. F. (1985).* The effect of VDU text-presentation rate on
reading comprehension and reading speed. ACM SIGCHI Bulletin, 16(4), 1-6.
Trauzettel-Klosinski, S., & Dietz, K. (2012).* Standardized assessment of reading performance: The new
international reading speed texts IReST. Investigative Ophthalmology & Visual Science, 53(9), 5452-5461.
Traxler, A. E. (1934). The relation between rate of reading and speed of association. Journal of
Educational Psychology, 25(5), 357-365.
Tyrrell, R. A., Pasquale, T. B., Aten, T., & Francis, E. L. (2001).* 47.4: Empirical Evaluation of User
Responses to Reading Text Rendered Using ClearType™ Technologies. In SID Symposium Digest of
Technical Papers (Vol. 32, No. 1, pp. 1205-1207). Oxford, UK: Blackwell Publishing Ltd.
Vajjala, S., & Meurers, D. (2014). Readability assessment for text simplification: From analysing
documents to identifying sentential simplifications. ITL-International Journal of Applied Linguistics,
Van Landingham, S. W., West, S. K., Akpek, E. K., Muñoz, B., & Ramulu, P. Y. (2014).* Impact of dry eye
on reading in a population-based sample of the elderly: the Salisbury Eye Evaluation. British Journal of
Ophthalmology, 98(5), 639-644.
Van Orden, G. C., Johnston, J. C., & Hale, B. L. (1988). Word identification in reading proceeds from
spelling to sound to meaning. Journal of Experimental Psychology: Learning, Memory, and Cognition,
Veldre, A., & Andrews, S. (2014).* Lexical quality and eye movements: Individual differences in the
perceptual span of skilled adult readers. The Quarterly Journal of Experimental Psychology, 67(4), 703-
Venkatagiri, H. S. (1999).* Clinical measurement of rate of reading and discourse in young adults. Journal
of Fluency Disorders, 24(3), 209-226.
Vienot, F., Durand, M. L., & Mahler, E. (2009).* Kruithof's rule revisited using LED illumination. Journal of
Modern Optics, 56(13), 1433-1446.
Wagenmakers, E. J., Love, J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., ... & Meerhoff, F. (2018).
Bayesian inference for psychology. Part II: Example applications with JASP. Psychonomic Bulletin &
Review, 25(1), 58-76.
Wagner, R. K., & Sternberg, R. J. (1987).* Executive control in reading comprehension. In B.K. Britton &
S.M. Glynn (Eds.), Executive control processes in reading (pp. 1-21). Routledge.
Wang, X., Sui, X., & White, S. J. (2019). Searching for a word in Chinese text: insights from eye movement
behaviour. Journal of Cognitive Psychology, 1-12. Advance publication DOI:
Ward, S.C. (2002) Modernizing the mind: Psychological knowledge and the remaking of society.
Westport, CT: Praeger.
Warmington, M., Stothard, S. E., & Snowling, M. J. (2013).* Assessing dyslexia in higher education: the
York adult assessment battery-revised. Journal of Research in Special Educational Needs, 13(1), 48-56.
Wetzel, R., & Knowlton, M. (2000).* A Comparison of Print and Braille Reading Rates on Three Reading
Tasks. Journal of Visual Impairment & Blindness, 94(3), 146-154.
Whimbey, A., & Lochhead, J. (1999). Problem solving and comprehension (6th edition). Mahwah, NJ:
Whitford, V., & Titone, D. (2015).* Second-language experience modulates eye movements during first-
and second-language sentence reading: Evidence from a gaze-contingent moving window paradigm.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 41(4), 1118-1129.
Wooster, G. F. (1954).* An experimental study of the reading-rate controller. Journal of Educational
Psychology, 45(7), 421-426.
Wu, H. C., Lee, C. L., & Lin, C. T. (2007).* Ergonomic evaluation of three popular Chinese e-book displays
for prolonged reading. International Journal of Industrial Ergonomics, 37(9-10), 761-770.
Yan, M., Kliegl, R., Richter, E. M., Nuthmann, A., & Shu, H. (2010). Flexible saccade-target selection in
Chinese reading. The Quarterly Journal of Experimental Psychology, 63(4), 705-725.
Yan, M., Zhou, W., Shu, H., & Kliegl, R. (2015).* Perceptual span depends on font size during the reading
of Chinese sentences. Journal of Experimental Psychology: Learning, Memory, and Cognition, 41(1), 209-
Yap, M. J., Liow, S. J. R., Jalil, S. B., & Faizal, S. S. B. (2010). The Malay Lexicon Project: A database of
lexical statistics for 9,592 words. Behavior Research Methods, 42(4), 992-1003.
Yaworski, J. (2005). Getting ahead: Fundamentals of college reading. Pearson/Longman.
Yen, N. S., Tsai, J. L., Chen, P. L., Lin, H. Y., & Chen, A. L. (2011).* Effects of typographic variables on eye-
movement measures in reading Chinese from a screen. Behaviour & Information Technology, 30(6), 797-
Yuan, J., Liberman, M., & Cieri, C. (2006). Towards an integrated understanding of speaking rate in
conversation. In Ninth International Conference on Spoken Language Processing. Available at
Zagoruiko, N. G., & Tambovtsev, Y. (1982). Aspects of human performance in an intensive speech task.
International Journal of Man-Machine Studies, 16(2), 173-181.
Zambarbieri, D., & Carniglia, E. (2012).* Eye movement analysis of reading from computer displays,
eReaders and printed books. Ophthalmic and Physiological Optics, 32(5), 390-396.
Zang, C., Zhang, M., Bai, X., Yan, G., Paterson, K. B., & Liversedge, S. P. (2016). Effects of word frequency
and visual complexity on eye movements of young and older Chinese readers. The Quarterly Journal of
Experimental Psychology, 69(7), 1409-1425.
Zeng, Y., Bai, X., Xu, J., & He, C. G. H. (2016).* The Influence of E-book Format and Reading Device on
Users’ Reading Experience: A Case Study of Graduate Students. Publishing Research Quarterly, 32(4),
Zeri, F., Naroo, S. A., Zoccolotti, P., & De Luca, M. (2018).* Pattern of reading eye movements during
monovision contact lens wear in presbyopes. Scientific Reports, 8(1), 15574.
Zhang, S., & Perfetti, C. A. (1993).* The tongue-twister effect in reading Chinese. Journal of experimental
psychology: Learning, Memory, and Cognition, 19(5), 1082-1093.
Zhao, J., Kwok, R. K., Liu, M., Liu, H., & Huang, C. (2017).* Underlying skills of oral and silent reading
fluency in chinese: perspective of visual rapid processing. Frontiers in Psychology, 7, 2082.
Zutell, J., & Rasinski, T. (1989). Reading and spelling connections in third and fifth grade students.
Reading Psychology: An International Quarterly, 10(2), 137-155.
Zwaan, R. A. (1991).* Some parameters of literary and news comprehension: Effects of discourse-type
perspective on reading rate and surface structure representation. Poetics, 20(2), 139-156.
Table 1: Studies investigating silent reading in languages with Latin alphabet
Aberson & Bouwhuis (1997)
General population with
Acklin & Papesh (2017)
Altaribba et al. (1996, Exp 1)
Ambrosino et al. (1974)
Arnell et al. (2009)
Ashby et al. (2012)
Baddeley et al. (1985, Exp 1)
Members of the unit
Baddeley et al. (1985, Exp 2)
Members of the unit
Hourcade (2011, Exp 1)
Ball & Hourcade (2011, Exp 2)
Bassin & Martin (1976)
Bear & Imus (1938)
Bell & Perfetti (1994)
Bell & Perfetti (1994)
et al. (2012)
Bellows & Rush (1952)
Benedetto et al. (2014)
Benedetto et al. (2015)
Benevides & Peterson (2010)
Benton et al. (1984)
Boije & Gustafsson (2012)
Bowers & Reid (1997)
Bowman et al. (2010)
Bridgeman & Montegut (1993)
Britton et al. (1978, Exp 1)
Burt & Fury (2000)
Callens et al. (2012)
Calvo et al. (1994, Study 1+2)
Calvo et al. (1994, Study 3)
Campbell et al. (1989, Exp 1+2)
Choi et al. (2015)
et al. (2017)
Cohen & Waiss (1991)
Conlon & Sanders (2011)
Cupples & Holmes (1992, Exp 1)
Cupples & Holmes (1992, Exp 2)
Deacon et al. (2012)
Dirix et al. (2019)
Dixon et al. (1988)
Dumler & Dumler (1958)
Dwyer & West (1994)
Dyson & Haselgrove (2000)
Dyson & Haselgrove (2001)
Dyson & Kipping (1997)
Everatt & Underwood (1994)
Fisher (1975, Exp 1)
Franken et al. (2015)
Freeburne & Fleischer (1952)
Glimne et al. (2015)
Graf & Levy (1984, Exp 2 & 3)
Grisham et al. (1993)
Gunraj & Klin (2012, Exp 1 & 3)
Gunraj et al. (2014, Exp 2 & 4)
Hartley (1993, Exp 2)
Students and staff
Hartley et al. (1994)
Haught & Walls (2002)
Hebert (2017, Exp 1)
Hebert et al. (2018)
Heij & van der Meij
Henry & Lauer (1939)
Henry et al. (2018)
Henry et al. (2018)
Hess & Tate (1992)
Hunt et al. (1981)
Hyona & Niemi (1990, Exp 1)
Hyona & Niemi (1990, Exp 2)
Hyona et al. (2002)
Jackson & Balota (2012, Exp 4)
Jackson et al. (1999)
Jensen et al. (1972)
Johansson et al. (2014)
Johnson et al. (2018)
Jones et al. (2012)
Juola et al. (1982, Exp
Just & Carpenter (1987)
Kamienkowski et al. (2016)
Karakus et al. (2018)
King et al. (1969)
Kingston & George (1957)
Kintsch & Monk (1972, Exp 1)
Kirby et al. (2008)
Korinth & Fiebach (2018)
Kruk & Muter (1984, Exp 1)
Kuperman et al. (2019)
Laubrock & Kliegel (2015)
Lewandowski et al. (2003)
Liversedge et al. (2016)
Liversedge et al. (2016)
Lloyd & McKelvie (1992)
Mackensen & Stichler (1963)
Mackensen & Stichler (1963)
Mackensen & Stichler (1963)
Mackensen & Stichler (1963)
Mackensen & Stichler (1963)
Mackensen & Stichler (1963)
Mackensen & Stichler (1963)
Mak & Willems (2019)
Maki et al. (1994)
Chang & Gould (2008)
Masson (1982, Exp. 1)
Masterson & Hayes (2004)
Mathews et al. (2017)
general population (older)
Maxwell & Mueller (1965)
Mayr et al. (2017, Exp 1)
McConkie & Meyer (1974)
McConkie & Rayner (1974)
McConkie & Rayner (1974)
McConkie et al. (1973, Exp 1)
McConkie et al. (1973, Exp 2)
Miller & Coleman (1972)
Monk (1984, Exp 1
Moys et al. (2019)
Muter & Maurutto (1991, Exp 1+2)
Muter et al. (1982)
Noyes & Garland (2003)
Oliver et al. (2005)
Oquist & Goldstein (2003)
Pashler et al. (2013, Exp 1)
Paterson & Jordan (2010)
Perea & Acha (2009)
Perrin et al. (2014)
general population with
Preston & Botel (1952)
Preston & Tuft (1948)
Ramulu et al. (2013)
Rayner (1986, Exp 1
Rayner et al. (1998,
Rayner et al. (2010)
Rello & Baeza
Ricciardi & Di Nocera (2017, Exp
Roberts et al. (2013, Exp 1+2)
Rose & Rostas (1947, Exp 4)
Sackstein et al. (2015)
Samuels & Dahl (1975, Exp 2)
Sekey & Tietz (1982, Exp 1+2)