Children’s Eye Movements during Reading.
Hazel I. Blythea and Holly S.S.L. Josephb*
a School of Psychology
University of Southampton
b Department of Experimental Psychology
University of Oxford
South Parks Road
*The order of authors is alphabetical, and entirely arbitrary
Word count = 10,010
Keywords: CHILDREN; EYE MOVEMENTS; READING; DEVELOPMENT
Children’s eye movements during reading. In this chapter, we evaluate the literature on
children’s eye movements during reading to date. We describe the basic, developmental changes
that occur in eye movement behaviour during reading, discuss age-related changes in the extent
and time course of information extraction during fixations in reading, and compare the effects of
visual and linguistic manipulations in the text on children’s eye movement behaviour in relation
to skilled adult readers. We argue that future research will benefit from examining how eye
movement behaviour during reading develops in relation to language and literacy skills, and use
of computational modelling with children’s eye movement data may improve our understanding
of the mechanisms that underlie the progression from beginning to skilled reader.
There has been a great deal of research that has used eye movement recordings to examine
the psychological processes underlying skilled adult reading (see Rayner, 1998, 2009 for a
review). This line of research has grown to the point where there are several well-developed
computational models that can account for many of the phenomena associated with eye
movement behaviour during reading (e.g., the SWIFT model, Engbert, Longtin, & Kliegl, 2002;
Engbert, Nuthmann, Richter, & Kliegl, 2005; the E-Z Reader model, Pollatsek, Reichle, &
Rayner, 2006; Reichle, Pollatsek, Fisher, & Rayner, 1998; Reichle, Pollatsek, & Rayner, 2006;
Reichle, Rayner, & Pollatsek, 1999, 2003; see also the chapters by Reichle, and by Engbert &
Kliegl in the present volume). These models are, however, currently limited to explaining the
end point of development: the skilled adult reader. This is largely a consequence of there having
been very few studies that have examined children’s eye movements during reading in
proportion to the vast body of research that has studied adults.
It is unsurprising that there has been little research in this area with children. Taking
accurate eye movement recordings is a process that requires the participant to sit very still and
follow instructions from the researcher, often for prolonged periods of time, requiring a lot of
patience. This is difficult with adults and, understandably, even more so with children as
participants. Finding young children who are willing and able to cooperate for the duration of an
experiment can be extremely challenging and this has been a constraining factor in the sample
size of some studies (depending on the particular equipment used and, consequently, the spatial
and temporal accuracy of the data). As eye tracking technology improves, however, it is
increasingly possible to take accurate eye movement recordings from larger samples of children
across a broader range of ages.
In order to fully understand the cognitive processes underlying reading it is necessary to
study the trajectory of reading development through on-line research methods such as eye
movement recordings. In this chapter, we review the small, existing literature on children’s eye
movements during reading both to assess the current state of knowledge in this field, and also in
an effort to highlight areas where future eye movement research might be effective in improving
understanding of the cognitive changes underlying children’s reading development. We do not
discuss in detail the body of literature that exists on the eye movements during reading of
children with dyslexia, as it is beyond the scope of this chapter. In experiments investigating the
relationship between dyslexia and eye movements, typically developing children are frequently
used for both reading age and chronological age control groups; however, the data analyses are
strongly focused upon comparisons of the different groups and the experiments are not designed
to examine the characteristics of typically developing children’s oculomotor control during
reading. For a review of the literature on eye movements and dyslexia, see Kirkby, Webster,
Blythe and Liversedge (2008).
In this chapter, we first comment on some theoretical and methodological issues which are
particularly relevant to conducting eye movement research with children. Second, we discuss
several studies which were largely exploratory and which examined the basic characteristics of
children’s eye movements during reading at different ages. The third section considers
developmental changes in the encoding of visual information during fixations in reading. The
latter half of the chapter then focuses on the influences of specific manipulations within text
upon children’s cognitive processing, as indexed by their eye movement behaviour. We give a
review of studies which have manipulated lexical characteristics in text and then, following on
from this, the influence of higher-level processing manipulations are considered. The chapter
concludes with a discussion of how future research into children’s eye movements during
reading may inform our understanding of the cognitive developmental changes underlying the
progression from beginning to skilled reader.
2. Conducting eye movement research to investigate children’s reading
2.1. Theoretical issues.
First, it is vital to consider the possible contributions of chronological age and reading skill in
relation to each other, and how they might impact on eye movement behaviour. As age
increases, reading ability also improves as the child progresses from a beginning to a skilled
reader. Thus, any changes in eye movement behaviour that are observed between different age
groups may be attributable to the difference in chronological age, or they may be attributable to a
difference in reading skill. When a skilled adult reads a line of text, their eyes make a sequence
of extremely quick saccades that accurately position the eyes such that new information falls on
the fovea. As with many different motor skills, it is possible that control of these eye movements
improves with age in terms of speed and accuracy. Indeed, previous research using non-reading
tasks has demonstrated developmental decreases in saccadic latency (e.g., Cohen & Ross, 1978;
Groll & Ross, 1982; Huestegge, Radach, Corbic, & Huestegge, 2009) although saccadic
accuracy, peak saccadic velocity, and saccadic overshoot have all been observed to be the same
in children as in adults (e.g., Cohen & Ross, 1978; Fukushima, Hatta, & Fukushima, 2000;
Salman et al., 2006). Thus, one possibility is that on a task such as reading, where optimal
performance depends on the execution of small but fast and accurate eye movements,
improvements in motor control associated with increasing chronological age may be one factor
underlying the observed developmental changes in eye movement behaviour.
On the other hand, and perhaps more obviously, developmental changes in eye movement
control are also a behavioural consequence of increasing reading skill. As reading skill
increases, changes in eye movements will reflect the decrease in cognitive processing difficulty
associated with text comprehension. In this case, we might expect to see differences between
children of the same chronological age who vary significantly in reading skill. While several of
the studies reviewed in this chapter have been careful to exclude children who were not within
the normal range of reading ability for their age, to our knowledge no study has explicitly
addressed this potential dichotomy. The possibility that differences in eye movement behaviour
during reading between children and adults may be a combination of both age-related changes in
motor skill and differences in reading proficiency should also be considered. The relative
contributions of these two factors are not yet well understood in the eye movement literature;
however, future research may allow separate examination of the causes underlying age-related
change by including multiple child participant groups and matching both for chronological age
and reading age, as well as making systematic manipulations of text difficulty in relation to
reading skill (see discussion in Section 2.2 below).
Second, a critical issue with respect to age-related changes in eye movement behaviour
during reading is that of cause and effect. Does increased processing difficulty cause readers to
make more, longer fixations, etc., or is it the case that poor control of eye movements leads to
difficulty in reading? The results of two eye movement studies with children have indicated that
processing difficulty during reading causes changes in eye movement behaviour. Rayner (1986)
found that, within a group of 9- to 10-year old children, reading rates were slower, more
fixations were made, fixation durations were longer, and the perceptual span was smaller when
they read more difficult sentences compared to when they read relatively easy sentences. Thus,
for an individual reader, the difficulty of the reading material impacts upon eye movement
behaviour. Häikiö, Bertram, Hyönä and Niemi (2009) found that less able readers (aged 8-, 10-,
and 12-years as well as adults) have slower reading speeds and a smaller perceptual span than
more able readers (see Section 5.1 for an explanation of the perceptual span). Thus, processing
difficulty is reflected in the reader’s eye movement behaviour. It is less clear whether poor eye
movement control can cause reading difficulty. Compared to their typically developing peers,
children with dyslexia have been reported to have unstable fixation, increased numbers of
express saccades, poor visual attention span, and limited perceptual span (see Kirkby et al., 2008
for a review). Most researchers agree, however, that differences in eye movement behaviour are
either correlates or consequences of the reading difficulties experienced by children with
dyslexia (e.g., Rayner, 1985). There are no published data, to date, which provide compelling
evidence for a causal role of poor eye movement control in dyslexia.
Third, to reiterate, there are several well-developed computational models of eye movements
during reading but they are based solely on data from skilled adult readers, (Engbert et al., 2002;
Engbert et al., 2005; Pollatsek et al., 2006; Reichle et al., 1998; Reichle et al., 1999, 2003; Reilly
& Radach, 2006). Recent work using artificial learning agents has shown how eye movements
resembling those of a skilled adult reader can emerge as a consequence of reinforcement learning
(Liu & Reichle, 2010; Reichle & Laurent, 2006). No a priori assumptions were made about the
mechanisms underlying eye movement control; the agents were simply limited by constraints
relating to factors known to impact on eye movements during reading (time taken to program
saccades, word length as an index of processing difficulty associated with lexical identification,
visual acuity limitations on lexical identification, and distribution of attention across words in a
sentence). The agent was rewarded for each word that was identified, and was punished for time
spent on word identification – the goal was that all the words in the sentence should be identified
as quickly as possible. The behaviours that emerged were similar to those observed in skilled
adult reading. The agent learned to target saccades toward the centres of words, and decisions
about when (reading times) and where (refixation and word skipping probabilities) to move the
eyes were affected by the processing difficulty associated with a word. In the future, modelling
eye movement data collected from children as they read will provide the opportunity to
understand the mechanisms underlying developmental changes in eye movement behaviour
during reading, such as vision, language, attention, and basic oculomotor control, and
interactions between such mechanisms during the progression from beginning to skilled reader.
2.2. Methodological issues.
There are several methodological issues that should be considered in relation to
developmental studies of eye movements during reading. The first issue is that of control
groups, and this relates to the theoretical issue outlined above of how both chronological age and
reading skill might contribute to the development of eye movement behaviour during reading. In
the small number of studies which have used different groups of children in order to examine
age-related changes in reading performance, groups have generally been split by chronological
age. While this has certainly proven fruitful, it is worth noting that in other areas of
developmental research (such as investigation of the language and literacy difficulties associated
with developmental disorders), comparison groups are often controlled both for chronological
age and for cognitive ability or mental age (e.g., Bishop & Snowling, 2004; Laing, Hulme,
Grant, & Karmiloff-Smith, 2001; Leonard, 1998; Snowling, 2000). It is important that research
is undertaken to directly address the relative contributions of age and reading skill to the
development of eye movement behaviour during reading. Until this is better understood, the
inclusion of control groups for both chronological age and reading age would be helpful to avoid
inadvertent confounding of one with the other.
A second methodological issue worth noting is the importance of the use of age-appropriate
linguistic stimuli. One factor distinguishing between the studies reviewed in this chapter is their
selection of stimuli for readers of different ages. Some studies have presented each group with
age-appropriate stimuli. In this case, processing difficulty associated with reading can be
equated across the groups (assuming careful stimulus selection), but the analysis of age group
effects will necessarily be between-items; differences between groups may be attributable to
their age, or may be attributable to the particular set of sentences that was read by each group. In
other studies, participants of all ages read the same stimuli. In this latter case, analyses of age
group effects are within-items, thus eliminating the concern that any effects observed are related
to differences in materials rather than reflecting genuine effects of age group. There is, however,
the concern that any set of reading materials will be inherently more difficult for younger readers
compared to older readers. It is well-documented that for skilled adult readers, increased
processing difficulty impacts on eye movement behaviour. For example, word reading times are
longer on low frequency words than high frequency words (e.g., Henderson & Ferreira, 1990),
and reading times are longer and more regressions are made for sentences that are syntactically
complex compared to those that are simple (Warren, White, & Reichle, 2009); for a summary see
Rayner (1998, pp. 389-392). Clearly, having participants of different ages read the same
materials will lead to differences in eye movement behaviour that are, at least partially,
attributable to the processing difficulty associated with the text for readers of different ages.
A related point is that manipulations of text difficulty may affect readers differentially
depending upon their reading skill and experience of processing printed text. Factors known to
have a robust effect on adults’ eye movements during reading, such as the length or frequency of
a word, need to be carefully considered before being incorporated into experimental designs with
children. For example, there are many words which are low frequency for adult readers but high
frequency for children (e.g., dragon) and frequency of encounter, together with cumulative
frequency (the total number of times a word is encountered over time) and Age-of-Acquisition
(the age at which a word was acquired), may impact differently on lexical retrieval processes in
adults and children. Ideally, future research would control for age-appropriateness by
constructing multiple sets of experimental sentences, each designed for a different age group
(e.g., Häikiö et al., 2009) and, if possible, by using extensive pre-screening in order to compose
sentences that are roughly equivalent in processing difficulty for all participant groups.
While creating stimuli that include linguistic processing manipulations but are also
comparable in terms of processing difficulty across different age groups is extremely
challenging, in certain circumstances it may be possible. For example, it is possible to set up a
frequency manipulation in which items are high frequency for both adults and children of
different ages. In addition, context can be made particularly relevant to children (e.g., sentences
refer to well-known children’s characters, school etc.), thereby reducing (although not
eliminating) the gap between how easy adults and children will find sentences to process. In
addition, stimuli can pre-screened for variables such as comprehension, plausibility, and
predictability. Experimenters can attempt to equate these across age groups by pre-screening
more items than are needed and selecting a subset for the main experiment that are matched on
these variables between adults and children.
Third, there is a methodological issue which inevitably arises when conducting research with
children – increased variability in data. This usually exists not only in eye movement measures,
but also in related measures such as reading ability, and can lead to problems with statistical
analyses (e.g., violating the assumption of equal variance across groups for ANOVAs) as well as
making it difficult to obtain reliable effects. One possible solution is to use large sample sizes.
Alternatively, child participants may be grouped more closely with respect to either
chronological age or reading skill. Both solutions are increasingly possible as new eye tracking
devices make it far easier to test large numbers of children. Finally the increasing use of Linear
Mixed Effects (LME) analyses in eye movement research, in which both fixed and random
effects are accounted for, may reduce the problems caused by this increased variance in child
data. It may, however, also be the case that it is not realistic to expect reliable differences when
experimental manipulations which generate only small effects are employed. Certainly, eye
movement research into children’s reading is at an early stage compared to the advanced
literature on skilled adult readers where increasingly complex manipulations are employed to
investigate relatively small effects. Currently, the basic questions that need addressing with
respect to children’s eye movements during reading allow studies to employ strong
manipulations that generate large differences between groups of children. In the future,
however, the issue of increased variability in children’s data may be addressed by more sensitive
approaches to the selective recruitment and screening of child participants, together with careful
design and screening of experimental stimuli.
3. Basic characteristics of children’s eye movements during reading
For comparative purposes, means for different eye movement measures at different ages
from all of the published studies to date of typically developing children’s eye movements during
reading are shown in Table 1. The age-related changes are quite clear; as chronological age
increases, sentence reading times and fixation durations decrease, saccade amplitudes increase,
fewer fixations and regressions are made, refixation probability decreases and word skipping
probability increases (Blythe, Häikiö, Bertam, Liversedge, & Hyönä, 2010; Blythe et al., 2006;
Blythe, Liversedge, Joseph, White, & Rayner, 2009; Buswell, 1922; Häikiö et al., 2009;
Huestegge et al., 2009; Joseph, Liversedge, Blythe, White, & Rayner, 2009; McConkie et al.,
1991; Rayner, 1986; Taylor, 1965; Taylor, Frackenpohl, & Pettee, 1960). While all studies show
these same developmental changes in eye movement behaviour, they vary in their selection of
stimuli for readers of different ages. Some studies have used different materials for readers of
different ages in an effort to match reading difficulty across groups (Blythe et al., 2006;
McConkie et al., 1991; Taylor, 1965). In contrast, other studies have presented readers of all
ages with the same sentences in order to avoid differences in material as a confounding variable
between groups (Blythe et al., 2010; Blythe et al., 2009; Buswell, 1922; Häikiö et al., 2009;
Huestegge et al., 2009; Joseph et al., 2008; Joseph et al., 2009; Rayner, 1986). In this second
group of studies, the stimuli were typically aimed at the level of the youngest readers tested and,
as a consequence, were very easy to read for the adult participants.
Given their differences in age groupings of participants and the manner in which certain
measures are reported (e.g., the number of regressions being reported either as the number made
per sentence or as a percentage of all movements made), it is not possible to make meaningful
comparisons between the magnitude of effects observed in studies that have used the same
stimuli for readers of all ages and the magnitude of effects observed in studies that have used
different stimuli for each age group. Importantly, though, all studies show the same basic
developmental trends – it does not seem to be the case that the selection of more or less difficult
materials for any given age group will differentially reflect age-related changes in eye movement
behaviour to a substantive degree. Further, the results of these studies broadly agree that
developmental changes in eye movement behaviour reach adult levels around the age of 11-years
(see Table 1).
Insert Table 1 about here
One interesting point to note is that, in studies where different sentences were written for
each age group in order to try and control processing difficulty, age-related changes in eye
movement behaviour may be a consequence of age-related changes in identification of the
individual words in the sentences. Specifically, these differences in eye movement behaviour
may reflect slower or less efficient lexical identification in children compared to adults, despite
the sentences being age-appropriate. This suggestion is supported by more recent data showing
that linguistic skills (word/ picture naming speed) but not oculomotor skills (saccade latency on
pro- and anti-saccade tasks) at the age of 8 predict sentence reading times at the age of 10
(Huestegge et al., 2009). To be clear, these data indicate that a child’s lexical processing ability
can predict some aspects of their eye movement behaviour across sentences as wholes. The most
important point to note, however, is that basic developmental changes in eye movement
behaviour when reading sentences (as shown in Table 1) have been observed very consistently in
all the studies published to date.
4. Saccadic targeting to words
An important aspect of eye movement behaviour during reading is where a reader locates
their initial fixation (or subsequent refixation) on a word, and the consequences that landing
positions on words have for ongoing language processing. This is important because we know
that in adult readers, initial fixation location influences how quickly a word is processed (Vitu,
McConkie, Kerr, & O'Regan, 2001), and also how likely it is that the word will be refixated
(McConkie, Kerr, Reddix, & Zola, 1988; McConkie, Kerr, Reddix, Zola, & Jacobs, 1989).
Three studies have examined landing position effects in children during normal text reading
(Joseph et al., 2009; McConkie et al., 1991; Vitu et al., 2001).
McConkie et al. (1991) reported data from Grimes (1989), which showed that during their
first year of reading instruction, children exhibited the same pattern of landing positions as
adults, although full analyses were not reported. Furthermore, McConkie et al. found that, like
adults, children were more likely to refixate a (five-letter) word following an initial fixation on
the space before the word or on the first letter, than if the first fixation was close to the word
centre (although these inferences were made from observing trends in the data rather than from
conducting formal statistical analyses).
Vitu et al. (2001) conducted extensive analyses on three data sets, one of which used data
from children who were approximately 12-years-old. Although the aim of the study was not to
compare children and adults, the data from the Vitu et al. study showed that children did not
appear to differ from adults in the locations of their first or second fixations. Like adults,
children targeted their saccades towards the word centre and landing position distributions
shifted with word length (that is, the longer the word, the further into the word children’s
fixations were located). Also, like adults, and similar to the data reported by McConkie et al.,
children were more likely to refixate a word following an initial fixation away from the word
centre (i.e., at the beginning or end of a word) than following a fixation close to the word centre.
Finally, Joseph et al. (2009) used tightly controlled experimental materials to make direct
comparisons of landing positions between adults and children aged 7- to 11-years, as they read
sentences containing target words which were four, six or eight letters long. They found no
reliable differences in the location of initial fixations between adults and children for any word
length. They also found, in line with previous studies, that both adults and children were more
likely to refixate a word following an initial fixation away from the word centre. In addition,
they found that children’s refixation saccades were smaller than those of adults, and tended to be
regressive more often than adults’, although this trend was not significant.
Taken together, the data from these three studies show that very early in reading
development (as young as 7-years), children target their saccades towards the word centre
(McConkie et al., 1991; Vitu et al., 2001), and they do not differ reliably from adults in their
initial landing positions (Joseph et al., 2009). Furthermore, children (like adults) are more likely
to refixate a word following an initial fixation away from the word centre (that is at the
beginning or end of a word) than following a fixation close to the word centre (Joseph et al.,
2009; McConkie et al., 1991; Vitu et al., 2001), presumably because their initial fixation location
does not allow them to extract the visual information necessary to complete lexical identification
of the word. However, children appear to be less efficient than adults in targeting their refixation
saccades (Joseph et al., 2009).
It seems, then, that although children are limited in the amount of parafoveal information
available to them during reading compared to adults (i.e., they have smaller perceptual spans: see
Section 5.1 below), the parafoveal information that is available to them during normal text
reading is used effectively to guide their oculomotor behaviour in order to maximise reading
efficiency (although they appear to lag behind adult efficiency when more than one fixation on a
word is required). It is worth noting that some aspects of eye movement behaviour during
reading may arise due to basic oculomotor phenomena, independent of learning to read. While
reading development might be linked to deciding which word to fixate, the actual saccade
targeting and preview mechanisms may simply be a basic characteristic of the eye movement
control; for example, the finding that both adults and children both tend to target saccades to the
middle of words may be characterised as a global effect (Findlay, 1982), a phenomenon not
necessarily associated with reading. Nevertheless, research to date shows that children target
their saccades to word centres, and this ability develops either before reading instruction begins
or else within the first year of reading instruction.
5. Extent and time course of information extraction during fixations in reading.
There have been several strands of research that have employed innovative methodologies to
examine in detail how visual characteristics of text affect children’s ongoing language
processing. This section reviews studies that have used the moving window technique, the
boundary paradigm, and the disappearing text paradigm in order to investigate changes in both
the spatial extent and the time course of information extraction during fixations with respect to
development. We will consider each of these lines of research in turn.
5.1. Spatial limitations on information extraction.
Two studies have examined developmental changes in the perceptual span, the area of text
around the point of fixation from which useful information can be extracted (Häikiö et al., 2009;
Rayner, 1986). Each of these will be discussed in turn, in some detail, as the ability to pre-
process information from words to the right of fixation is a hallmark component of skilled adult
reading, and being deprived of the opportunity to pre-process information from the right of
fixation is extremely detrimental to adults’ reading (Rayner, 1975; Rayner, Inhoff, Morrison,
Slowiaczek, & Bertera, 1981; Rayner, Liversedge, & White, 2006; Rayner, McConkie, & Zola,
1980; Rayner, Well, Pollatsek, & Bertera, 1982).
The perceptual span has been measured using the moving window paradigm, in which there
is an area of text around the point of fixation which is presented normally, and this window of
text moves with the eyes as the reader progresses through the sentence (McConkie & Rayner,
1975). Outside the window, the text is mutilated in some way – typically replacing the letters
with xs, or other, visually-similar letters. The size of the window of unmutilated text that is
available to the reader is manipulated, and reading behaviour for different window sizes is
compared to reading behaviour for text which is presented normally in its entirety. Small
windows typically reduce reading speed, as the reader is unable to pre-process information
outside the window. The window size at which reading speed becomes equal to that for normal
text shows the perceptual span. As the reader can read at their full speed with a window of a
particular size, it is inferred that they do not make use of information further from the point of
fixation than is available to them in that window. The perceptual span for adults is asymmetric
about the point of fixation, extending from the beginning of the fixated word to around 14 or 15
characters to the right (see Rayner, 1998 for a summary).
In 1986, Rayner published a paper on the development of the perceptual span during reading.
Four experiments were conducted that examined the size and asymmetry of the perceptual span
in children of different ages compared to skilled adult readers, as well as the influence of text
difficulty upon the perceptual span. The first experiment used windows that were symmetrical
about the point of fixation; the perceptual span was shown to extend 23 characters around the
point of fixation (11 characters to either side of fixation) for the two youngest groups of readers
(7- and 9-years). In contrast, for 11-year-old children and the adults the perceptual span was
larger, extending 29 characters around the point of fixation (or, 14 characters either side of
fixation). When the window size was determined in terms of words rather than characters, and
word spaces outside the window were preserved so that the reader had parafoveal access to
upcoming word length information, both reading rate and eye movement data again showed that
younger readers have a smaller perceptual span than older readers, with a span of one word to the
right of fixation for 7-year-olds but a span of two words to the right of fixation for 9-year-olds,
11-year-olds, and adults. Comparisons across word- and character-defined window conditions
also showed that, consistent with previous adult studies, readers as young as 7-years-old pre-
process word length information at a greater eccentricity from fixation than they pre-process
With respect to the symmetry of the perceptual span, reading was equally disrupted for
participants of all ages when the availability of information to the left of fixation was restricted;
the characteristic asymmetrical span for adults is present in children by the age of 7-years.
Finally, the data showed very clearly that the window size which allowed maximum reading
speed was smaller for more difficult sentences – the perceptual span was reduced when the
reader experienced greater processing difficulty. Thus, these experiments showed that several
different aspects that characterise the perceptual span in skilled adult readers are established in
readers as young as 7-years-old (the acquisition of gross properties of the upcoming text at a
greater eccentricity than more specific letter information, and the asymmetry of the perceptual
span). Alongside these similarities, some differences were also found. Compared to adults,
younger readers have a reduced perceptual span and proportionally more of their processing
capacity is devoted to the fixated word. Manipulations of text difficulty with the children
supported the argument that, when comparing adults and children, age-related changes in the
overall size of the perceptual span were, at least partially, attributable to differences in
processing difficulty. Note that increased processing difficulty has also been shown to reduce
parafoveal pre-processing in skilled adult readers (Henderson & Ferreira, 1990; White, Rayner,
& Liversedge, 2005).
Surprisingly, no research was conducted to extend or challenge these findings for 20 years.
A recent study has, however, looked in detail at the development of the letter identity span – the
eccentricity from fixation at which readers of different ages can access letter-specific information
(Häikiö et al., 2009). Rayner showed that access to some degree of word-specific information
was limited to either one or two words to the right of fixation depending on the reader’s age,
while word length information was processed at greater eccentricities. Häikiö et al. designed
their experiment to examine more closely the level of detail associated with individual letters that
readers can access during reading at varying eccentricities.
Letters outside the window were replaced by visually similar characters (i.e., replacing an o
with a c), with word boundaries preserved. In this way, readers were able to pre-process both
word length and letter feature information outside the moving window, but it was only inside the
window that correct letter identities were available. First, the data clearly showed developmental
change in the letter identity span. At 8-years, the letter identity span extends five characters to
the right of fixation, at 10-years it extends seven characters, and at 12-years and for adults it
extends nine characters to the right of fixation. These data were compared to those from
Rayner’s experiments and showed that, for all age groups, letter identity span was smallest, with
letter feature information available relatively further to the right, and with word length
information available further still from fixation. Interestingly, they also compared the letter
identity spans of fast and slow readers of the same age. Clear differences were found at all ages,
with slower readers having a smaller span than faster readers. These data again suggest that
reading skill, rather than chronological age, is responsible for developmental differences in the
Some recent work, also investigating children’s pre-processing of upcoming information in a
sentence, has specifically examined whether pre-processing is different between words compared
to within words (Häikiö, Bertram, & Hyönä, 2010). Häikiö et al. compared adults’ and 8-, 10-,
and 12-year-old children’s pre-processing of the second consituent of compound words (e.g., of
“boy” in “cowboy”) to their pre-processing of the second word in adjective-noun pairs (e.g., of
“bad boy”) using the boundary paradigm (Rayner, 1975). In this paradigm an invisible boundary
was placed before the target word/s. Prior to the reader making a saccade that crossed the
boundary, a preview letter string was presented in the target location; in this experiment, the
letters of the target word/s were replaced with visually similar letters. When a saccade crossed
the invisible boundary, the preview string changed to the target word. Since this change
occurred during a saccade, when visual input is suppressed, the change was not noticed by the
reader. In all age groups, pre-processing was greater within a compound word than between the
words in an adjective-noun pair (see also Juhasz, Pollatsek, Hyönä, Drieghe, & Rayner, 2009).
They also found a significant parafoveal-on-foveal effect within the compound words, but not in
the adjective-noun pairs, in both adults and children (for a full explanation and discussion of
parafoveal-on-foveal effects, see the chapter by Drieghe in the present volume). These data
show that from as young as 8-years, the allocation of attention during reading and, thus, pre-
processing of upcoming information is determined to a significant degree by the visual cues
corresponding to word boundaries (rather than simply extending certain number of character
spaces ahead, irrespective of whether this includes one or multiple words).
To summarise, developmental changes have been shown to occur in the pre-processing of
three different aspects of text within the perceptual span (word length, letter feature, and letter
identity) such that this information can be acquired further to the right from the point of fixation
as age increases. Furthermore, these age-related increases in the perceptual span are driven by
the underlying improvement in reading skill , reinforcing understanding of the perceptual span as
an index of the allocation of attention and processing resources during reading rather than simply
being a low-level perceptual restriction on eye movement behaviour. That said, the left-right
asymmetry of the perceptual span is established by the age of 7-years; clearly, relatively little
experience of processing printed text is necessary to develop one of the key characteristics of
visual information processing during skilled adult reading. This asymmetry reflects the
allocation of attention to upcoming words for parafoveal pre-processing. In our view, the
development of parafoveal pre-processing during sentence reading is an integral aspect of the
progression from beginning to skilled reader. The finding that the perceptual span is asymmetric
in 7-year-olds is remarkable, given that reading skills at this age are still relatively basic and
measures of eye movement behaviour during reading (such as fixation durations and sentence
reading times) continue to show developmental change for a further four years, on average,
before reaching adult levels of performance. To our knowledge, no research has been conducted
to examine developmental changes in sensitivity to specific characteristics of upcoming words in
parafoveal pre-processing, despite the large body of literature on this topic for adults (for a
review, see Rayner, 1998, 2009). Future research must address how parafoveal pre-processing
continues to develop with age and reading skill, as 7-year-olds are still a long way from being
Given that the perceptual span is related to the reader’s cognitive processing of the text, we
anticipate that as reading skills improve with age: (1) a beginning reader will become sensitive to
increasingly detailed information from the word to the right of fixation; (2) the extent to which
they pre-process information from the upcoming word will increase; (3) reading times on
directly fixated words will decrease as a consequence of greater parafoveal pre-processing of
those words. Through such mechanisms we believe that the process of lexical identification will
become quicker and more efficient with age, not only due to more efficient linguistic processing
during direct fixation but also, at least partially, due to developmental changes in parafoveal pre-
processing. Increased parafoveal pre-processing of upcoming words in sentences will decrease
the demand on processing resources necessary to identify that word during direct fixation; this in
turn will facilitate parafoveal pre-processing of the next word. This cycle of pre-processing a
word and subsequent facilitation of lexical identification during direct fixation will develop with
reading skill, and it seems likely that this will be reflected in shorter fixation durations, fewer
refixations, and increased word skipping probabilities – an increasingly mature pattern of eye
movement behaviour during reading.
5.2. Temporal limitations on visual information extraction.
Recent research has compared children of different ages with adults using the disappearing
text paradigm (Liversedge et al., 2004; Rayner, Liversedge, White, & Vergilino-Perez, 2003).
This is a gaze-contingent change method, where there are invisible boundaries placed between
all the words within the sentence. Each time the reader’s eye crosses a boundary, the newly
fixated word disappears after a specified delay while the previously fixated word reappears; there
is only one word missing from the sentence at any one moment, but it is the word being fixated
by the reader. Critically, a pre-specified delay can be manipulated between the reader’s eye
crossing the boundary and the word disappearing, thus restricting the reader’s opportunity to
visually encode the word to the initial period (typically 60 ms) of the first fixation on the word.
Work with skilled adult readers has demonstrated that they are able to read and understand
sentences normally when presented as disappearing text with the word being presented for 60 ms
from fixation onset (Liversedge et al., 2004; Rayner et al., 2003). Interestingly, word frequency
effects were found upon fixation durations in the disappearing text condition; even when the
word was no longer visible, the reader’s cognitive processing determined when they would move
their eyes onto the next word in the sentence.
An interesting question that arose from the disappearing text studies was whether children
are able to encode visual information during fixations in reading as quickly and efficiently as
adults. Beginning readers are, by nature, less familiar with the written forms of words than
skilled adult readers. It may, therefore, be the case that they require longer presentation
durations than adults to allow successful visual encoding and, consequently, the initiation of
normal linguistic processing. Blythe and colleagues compared younger children (7- to 9-years),
older children (10- to 11-years) and adults as they read sentences presented as disappearing text
(Blythe et al., 2009). The sentences contained a target word that was manipulated for word
frequency. In two experiments, four different presentation durations were used – 40 ms, 60 ms,
80 ms, and 120 ms disappearing text compared to normally-presented sentences. The results
showed that even by the age of 7-years, there was a minimal impact of the disappearing text
manipulation upon children’s eye movement behaviour with all presentation durations tested.
Effects of word frequency were found upon single and first fixation durations even in the
disappearing text conditions, providing strong evidence that even with very short periods of
visual input, children aged 7-years are able to initiate normal lexical identification processes.
These results are comparable to those obtained for adults.
In summary, the data from studies which have investigated visual information extraction
during reading shows that (1) the perceptual span for reading increases in size with chronological
age, up to 11-years and is related to reading skill/ processing difficulty; (2) the characteristic left-
right asymmetry of the perceptual span associated with skilled adult reading of English has
developed by the age of 7-years; (3) by the age of 7-years, children accurately target their
saccades close to the word centre, as adults do; and (4) the speed of visual information encoding
during reading does not increase significantly after the age of 7-years. Thus, while spatial
aspects of information encoding continue to develop up to 11-years, where a child initially
fixates a word and the speed with which a word is visually encoded are in place just a few years
after beginning formal reading instruction, or perhaps even before (see Section 4).
In the latter half of this review, we will consider how various characteristics of the material
being read can impact upon the eye movements of children compared to adults. Studies in this
area have used eye movements as an index of the moment-to-moment psychological processing
that underlies reading, to examine the process of development from beginning to skilled reader.
6. Word-based effects in children
Two of the most robust effects in the adult eye movement literature are those of word length
and word frequency; that is, adults look longer at long than short words (Just & Carpenter, 1980;
Rayner, Sereno, & Raney, 1996); and at low frequency than high frequency words (e.g.,
Henderson & Ferreira, 1990; Inhoff, 1984; Inhoff & Rayner, 1986; Just & Carpenter, 1980;
Rayner, 1977; Rayner & Duffy, 1986; Rayner et al., 2003; Rayner & Raney, 1996). Perhaps for
this reason, the majority of studies which have manipulated an aspect of text to examine its effect
on children’s eye movements during reading have focussed on precisely these two
6.1. Word length.
Four studies have manipulated word length to investigate whether these robust effects
(whereby long words are fixated more often and for longer than short words) observed in adult
readers are also present in children. Hyönä and Olson (1995) recorded the eye movements of
both dyslexic children (mean age = 14.4 years) and reading-age-matched controls (mean age =
10.5 years) as they read aloud texts which contained words that were, subsequent to data
collection, categorised as short (5-6 letters), medium (7-8 letters), or long (9-11 letters), resulting
in a very high number of target words and, hence, a very rich data set. The reading material was
set at a higher difficulty level than most participants’ level of word recognition in order that
some reading errors would be generated. Hyönä and Olson found a strong effect of word length
in both groups which was apparent in gaze durations (the sum of all fixations made on a word
before the eyes leave the word either to the right or to the left), number of first pass fixations (the
number of fixations made on a word before leaving that word to the right or left), second pass
reading time (the sum of all fixations made on a word after having left the word for at least one
fixation); and number of second pass fixations (the sum of all fixations associated with second
In a silent reading experiment, Joseph et al. (2009) took a different approach to that of Hyönä
and Olson, and manipulated word length prior to data collection, which meant that they were
able to control for word frequency and predictability in their target words (one per sentence).
This allowed them to ensure that any effects observed were due to word length alone rather than
being modulated other linguistic variables. Furthermore, the experimental sentences were
designed to be age-appropriate for the youngest of the children (aged 7-years), which meant that
they were relatively easy for the older children (aged 11-years) and adults.
Like Hyönä and Olson, they found reliably longer gaze durations, more refixations, and
longer total reading times on long than short words. They also found that both adults and
children skipped short words more often than long words. Importantly, most of these effects
were larger in children than adult readers, suggesting that not only do children experience an
increased processing load when reading long as compared to short words, but the increase in
word length has a more substantial effect on children’s ongoing lexical processing as compared
to that of adult readers. Furthermore, due to careful control of word frequency and predictability,
it is likely that this difference between adults and children was due to the demands of visually
encoding long as compared to short words, suggesting that children require more and/or longer
visual samples of long words in order to reach the point at which lexical identification can
Similarly, Huestegge et al. (2009) found that children aged 8- and 10-years old exhibited
longer gaze durations and total fixation times on long (6- to 9-letter) compared to short (4- to 5-
letter) words and, like the previous two studies, did not observe an effect of word length on
children’s first fixation durations. They also found an interaction between age group and word
length on refixation time (gaze duration minus first fixation duration) showing that the word
length effect was greater in younger compared to older children. These data from silent reading
experiments suggest, therefore, that younger readers need additional processing time on long
words compared to older readers, and that this need decreases with age.
A more recent experiment has directly investigated the question of why children make
multiple fixations on long words (Blythe et al., 2010). Children aged 8- to 9-years, 10- to 11-
years, and adults read sentences containing long (8-letter) or short (4-letter) words, that were
presented either normally or as disappearing text (for an explanation of the disappearing text
paradigm, see Section 5.2 of this chapter). The 8-9-year-old children made fewer refixations on
long target words when they were presented under disappearing text conditions compared to
normal conditions (resulting in shorter gaze durations). However, they subsequently made more
regressions back to long words in the disappearing text condition, leading to no overall
difference in total fixation times on those long words between the normal and disappearing text
conditions. Thus, these younger children adopted an eye movement strategy by which they
obtained a second visual sample on the long words without incurring any cost to overall
processing time on this word. Such effects were reduced in the older children, and were minimal
in the adults, indicating that while younger children do require a second visual sample on 8-letter
words, by the age of 10-years one visual sample is usually sufficient.
Together, these studies provide compelling evidence that children (up to the age of around 9-
years) are slower and less efficient at processing words, evidenced by their need for multiple and
longer visual samples when reading long compared to short words; this applies in the domains of
both silent and oral reading. Moreover, word length effects are found in text that is relatively
difficult (Hyönä & Olson, 1995) or easy (Blythe et al., 2010; Joseph et al., 2009) for those
children reading it, and the effects observed are more pronounced in younger readers compared
to older readers (Huestegge et al., 2009).
6.2. Word frequency.
Of the relatively small number of studies that have investigated word-based effects in
children, many have specifically manipulated word frequency (Blythe et al., 2006; Blythe et al.,
2009; Huestegge et al., 2009; Hyönä & Olson, 1995). Word frequency refers to how often a
word is encountered, as indexed by corpora such as CELEX (Baayan, Piepenbrock, & Gulikers,
1995) and Kučera and Francis (Kučera & Francis, 1967), which document how frequently a
given word appears in a range of written texts. It is likely that this particular variable was chosen
in all of the four studies because the word frequency effect is so robust in adult readers (e.g.,
Henderson & Ferreira, 1990; Inhoff, 1984; Just & Carpenter, 1980; Rayner & Duffy, 1986;
Rayner & Raney, 1996). It was, therefore, important to establish whether children’s eye
movements were influenced as immediately and reliably as those of adults by the frequency with
which a word is encountered.
In the same study outlined in the previous section, Hyönä and Olson (1995) also investigated
word frequency effects. Words of each length were categorised into three frequency groups:
low, medium and high. They found a strong effect of word frequency in oral reading which
could be seen in first fixation durations, showing that the frequency of a word has a very
immediate effect on processing in children, as it is known to do in adults. Importantly, the word
frequency counts of the target words were drawn from age-appropriate texts so that the high and
low frequency words were high or low frequency for children, rather than for adults. While this
study suggests that the frequency of occurrence of a word has an immediate effect on the reading
behaviour of children as well as adults, a more recent study by Blythe et al. (2006) failed to find
a frequency effect in children aged 7- to 11-years, despite finding a strong frequency effect in
their adult participants in the same study Importantly, this study used adult corpus data to index
The discrepancy between these two studies suggests that adult frequency counts may be
unsuitable for creating an effective manipulation with children. Blythe et al. (2009), however,
did find robust frequency effects in children aged 7- to 9-years and 10- to 11-years across two
separate experiments that used adult frequency counts in the target word manipulation (the
disappearing text study outlined in Section 5.2). These effects occurred in very early measures
of lexical identification, single and first fixation durations, in both children and adults. A third
study that used adult frequency counts also found effects on 8- to 10-year-old children’s eye
movements, but only in gaze durations and total reading times (Huestegge et al., 2009). There
was a numerical trend for longer first fixations on low frequency words compared to high
frequency words, but this was not statistically significant. Unfortunately this study did not
include a control group of adults, and so it is not possible to gauge how effective this
manipulation of frequency was for skilled adult readers in comparison to children.
Overall, word frequency appears to exert a strong influence on the ease of lexical
identification, as indexed by eye movement measures, from 7-years of age. It may, however, be
the case that while it is possible for adult counts of frequency to generate effects in children
(many words are, of course, high or low in frequency for adults and children alike), using
frequency counts from age-appropriate texts is a more reliable means of obtaining robust effects.
There is a clear need for a study to manipulate both adult and child frequencies and to examine
whether they differentially affect eye movement behaviour in readers of different ages. Future
research will also benefit from the careful control of potentially confounding linguistic variables
such as Age-of-Acquisition and predictability, as well as the age-appropriateness of the texts
used to index frequency, in order to establish more firmly how word frequency affects children’s
ongoing lexical processing. Finally, it should be noted that these data provide extremely strong
evidence for cognitive control of eye movements during reading in children as young as 7-years.
That is to say, by the age of 7-years, children’s reading has developed to a level where their
linguistic processing of the fixated word determines when they move their eyes on through the
7. Post-lexical processing
Only one study, to our knowledge, has used eye-movements during reading to examine
children’s processing during reading at a post-lexical level. Joseph et al. (2008) investigated
how children (aged 7- to 11-years) and adults processed implausible and anomalous thematic
relations during reading (see also the chapter by Warren in the current volume). In their
experiment, participants read sentences such as (1a-1c) below. The sentences described events
in which an individual performed an action with an instrument. In each case, the verb had three
thematic roles (see sentences 1 a-c): an agent (Robert), an instrument (trap, hook or radio) and a
patient / theme (mouse).
1a. Robert used a trap to catch the horrible mouse that was very scared.
1b. Robert used a hook to catch the horrible mouse that was very scared.
1c. Robert used a radio to play the horrible mouse that was very scared.
In all three sentences the instrument (trap, hook or radio) could be plausibly used in
conjunction with the main verb (catch or play). In the implausible condition (1b), however, the
patient (mouse) was incongruous as the object of the verb (catch) given the particular instrument
used (hook). That is, although hooks are often used to catch things (e.g., fish), and mice are
often caught, a hook is not often used to catch a mouse. By contrast, in the anomalous condition
(1c), the patient (mouse) could not be used in conjunction with the verb (in this case play).
Joseph et al. found that while children did not differ from adults in their anomaly detection
(both groups exhibited longer gaze durations on the critical word in the anomalous than control
conditions), they were delayed relative to adults in their implausibility detection. Adults showed
disruption in go-past time (the sum of all the temporally contiguous fixations from the first
fixation in a region until a fixation to the right of the region) in the post-target region, while
children only showed disruption in total time in the post-target region. Thus, there was a
difference in the time course of the disruption. The authors interpreted the data as showing that
children and adults were similar in terms of basic thematic assignment processes that occur
during reading, but that they differed in the efficiency with which they were able to integrate
pragmatic and real world knowledge into their discourse representation.
While it is difficult to make any firm conclusions based on data from a single study, the
results from Joseph et al. (2008) suggest that in some respects children may be adult-like in their
post-lexical written language comprehension, although they appear to be slower than adults to
incorporate real world knowledge information into their representation of sentence meaning.
Clearly, much more research is needed before this claim can be more than speculative. If
correct, however, this would suggest that early in reading development, children are very similar
to adults in terms of the oculomotor mechanisms they have in place in order to make the eye
movements necessary for written language comprehension, but that the efficiency and speed with
this these mechanisms and processes function is reduced relative to adult reading performance.
It may also be the case that there are developmental differences in how post-lexical processing
influences eye movement control.
8. Future directions
Overall, despite the small number of studies that have investigated children’s eye movements
during reading, we believe great progress has been made towards understanding oculomotor
control and on-line written language comprehension in children. This is an exciting time in the
field of developmental eye movement research into reading, as technological improvements
rapidly open up more possibilities for conducting well-controlled and innovative experiments
with children. There are several directions in which we anticipate the field will head in the
First, we believe that will be both timely and informative for new studies to examine the
relative contributions of variables such as chronological age, reading age, and IQ to the
development of oculomotor control during reading. From the existing literature on eye
movements during reading, the large variance in children’s data suggests that such factors are
influential. By recording the broader cognitive profiles of children with respect to their language
and literacy development, and analysing such information in relation to their eye movement data,
we hope that research will allow a more detailed understanding of the relationships between
development with age, reading skill and the moment-to-moment cognitive processing of text as
indexed by eye movements.
Second, we hope that the use of portable, highly-accurate eye tracking equipment will enable
well-controlled, longitudinal studies to be carried out. All of the research conducted in the field
thus far has been cross-sectional in design. While this work has proved extremely informative,
longitudinal studies will allow a more detailed examination of the developmental trajectory of
reading behaviour in relation to different aspects of linguistic processing. This approach is
already standard in studies using off-line measures of reading comprehension (e.g., Muter,
Hulme, Snowling, & Stevenson, 2004; Nation & Snowling, 2004). By adopting similar
experimental designs and using an on-line measure of cognitive processing as it occurs during
reading, researchers will be able to advance current understanding of literacy development by
taking a range of variables into account that differ within typically developing children.
Third, there has only been one published study which has examined linguistic effects beyond
the lexical level in children (Joseph et al., 2008). Further studies are needed to investigate
aspects of higher order linguistic processing (e.g., syntactic and semantic processing, and even
discourse processing) in children during sentence reading. While research thus far has shown
that adults and children from the age of 7-years are very similar in their processing of visual
linguistic stimuli at a lexical level, we consider it entirely possible that research examining post-
lexical processing will reveal developmental changes that continue beyond the age of 7-years.
One issue that may arise in relation to manipulations of higher order linguistic processing in
children is that of stimulus selection. It is not yet clear whether the use of higher-level linguistic
manipulations which have generated robust processing preferences in adult readers will produce
reliable effects in children. For example, given the increased variability introduced by including
children as participants, it may be that manipulations which result in small effect sizes in adult
readers (e.g., lexical ambiguity effects) will not produce detectable effects in children.
Fourth, one very informative development in the field would be for eye movement studies
with children to narrow the gap with the literature on literacy development based on studies that
have used off-line methods. There are many theoretical models of reading development that
have resulted from off-line research (see Ehri, 2005 for a review). We believe that, compared to
off-line studies, eye movement data will offer a more detailed understanding of the changes in
cognitive processing that underlie the progression from beginning to skilled reader and such data
will be extremely informative with respect to models of reading development. For example,
there are developmental changes in the extent to which children use letter, syllable, and word
representations as the access units in lexical identification; this varies across languages, and is
thought to reflect fundamental characteristics of the structure of the adult mental lexicon –
psycholinguistic grain size theory (Ziegler & Goswami, 2005). The sensitivity of eye movement
behaviour to cognitive processing during reading would make eye movement recordings an ideal
method with which to investigate such issues.
Finally, as discussed in Section 2.1, recent work has shown that artificial learning agents can,
with relatively few and simple constraints, develop eye movement behaviour resembling that of
skilled adult readers. In our view, the next step will be for the increasingly large body of
children’s data to be incorporated into the existing computational models of eye movements
during reading (e.g., the SWIFT model, Engbert et al., 2002; Engbert et al., 2005,; the E-Z
Reader model, Pollatsek et al., 2006; Reichle et al., 1998; Reichle et al., 2006; Reichle et al.,
1999, 2003; see the chapters by Reichle, and by Engbert & Kliegl in the present volume). To
reiterate, modelling children’s data in this way will facilitate our understanding of change in the
mechanisms (visual, attentional, linguistic, oculomotor control) underlying reading.
In summary, an increasing body of literature now exists which provides important,
preliminary data that document children’s basic oculomotor behaviour, lexical processing and, to
a limited extent, post-lexical processing during sentence reading. It is hoped that future research
will use these data as a benchmark to examine other populations such as younger children just
beginning the process of learning to read and children with developmental disorders, in particular
those with reading disabilities (e.g., dyslexia) or with more general language impairments (e.g.,
children with Specific Language Impairment or comprehension problems). In the long-term,
research in this area might go some way towards informing the design of interventions for
individuals who have difficulty learning to read. This may seem ambitious, but we believe that
there is every reason to be optimistic that eye movement research investigating children’s
reading will continue to flourish as it has started to over the last few years.
Baayan, H., Piepenbrock, R., & Gulikers, L. (1995). The CELEX lexical database (CD-ROM).
University of Pennsylvania, Philadelphia: Linguistic Data Consortium.
Bishop, D.V.M., & Snowling, M.J. (2004). Developmental dyslexia and specific language
impairment: Same or different? Psychological Bulletin, 130, 858-886.
Blythe, H. I., Häikiö, T., Bertam, R., Liversedge, S. P., & Hyönä, J. (2010). Reading
disappearing text: Why do children refixate words? Vision Research (in press).
Blythe, H. I., Liversedge, S. P., Joseph, H. S. S. L., White, S. J., Findlay, J. M., & Rayner, K.
(2006). The binocular co-ordination of eye movements during reading in children and adults.
Vision Research, 46, 3898-3908.
Blythe, H. I., Liversedge, S. P., Joseph, H. S. S. L., White, S. J., & Rayner, K. (2009). The
uptake of visual information during fixations in reading in children and adults. Vision
Research, 49, 1583-1591.
Buswell, G. T. (1922). Fundamental reading habits: A study of their development. Chicago:
University of Chicago Press.
Cohen, M. E., & Ross, L. E. (1978). Latency and accuracy characteristics of saccades and
corrective saccades in children and adults. Journal of Experimental Child Psychology, 26,
Ehri, L. C. (2005). Development of sight word reading: Phases and findings. In M. J. Snowling
& C. Hulme (Eds.), The science of reading: A handbook (pp. 135-154). Oxford: Blackwell.
Engbert, R., Longtin, A., & Kliegl, R. (2002). A dynamical model of saccade generation in
reading based on spatially distributed lexical processing. Vision Research, 42, 621-636.
Engbert, R., Nuthmann, A., Richter, E., & Kliegl, R. (2005). SWIFT: A dynamical model of
saccade generation during reading. Psychological Review, 112, 777-813.
Findlay, J. M. (1982). Global visual processing for saccadic eye movements. Vision Research,
Fukushima, J., Hatta, T., & Fukushima, K. (2000). Development of voluntary control of
saccadic eye movements: I. Age-related changes in normal children. Brain and
Development, 22, 173-180.
Grimes, J. (1989). Where first grade children look in words during reading. (Unpublished
master’s thesis). University of Illinois.
Groll, S. L., & Ross, L. E. (1982). Saccadic eye movements of children and adults to double-
step stimuli. Developmental Psychology, 18, 108-123.
Häikiö, T., Bertram, R., & Hyönä, J. (2010). Development of parafoveal processing within and
across words in reading: Evidence from the boundary paradigm. The Quarterly Journal of
Experimental Psychology (in press).
Häikiö, T., Bertram, R., Hyönä, J., & Niemi, P. (2009). Development of the letter identity span
in reading: Evidence from the eye movement moving window paradigm. Journal of
Experimental Child Psychology, 102, 167-181.
Henderson, J. M., & Ferreira, F. (1990). Effects of foveal processing difficulty on the perceptual
span in reading: Implications for attention and eye movement control. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 16, 417-429.
Huestegge, L., Radach, R., Corbic, D., & Huestegge, S. M. (2009). Oculomotor and linguistic
determinants of reading development: A longitudinal study. Vision Research, 49, 2948-
Hyönä, J., & Olson, R. K. (1995). Eye fixation patterns among dyslexic and normal readers:
effects of word-length and word-frequency. Journal of Experimental Psychology: Learning
Memory and Cognition, 21, 1430-1440.
Inhoff, A. W. (1984). Two Stages of Word Processing during Eye Fixations in the Reading of
Prose. Journal of Verbal Learning and Verbal Behavior, 23, 612-624.
Inhoff, A. W., & Rayner, K. (1986). Parafoveal word processing during eye fixations in reading:
effects of word frequency. Perception and Psychophysics, 40, 431-439.
Joseph, H. S. S. L., Liversedge, S. P., Blythe, H. I., White, S. J., Gathercole, S. E., & Rayner, K.
(2008). Children’s and adults’ processing of anomaly and implausibility during reading:
Evidence from eye movements. The Quarterly Journal of Experimental Psychology, 61,
Joseph, H. S. S. L., Liversedge, S. P., Blythe, H. I., White, S. J., & Rayner, K. (2009). Word
length and landing position effects during reading in children and adults. Vision Research,
Juhasz, B. J., Pollatsek, A., Hyönä, J., Drieghe, D., & Rayner, K. (2009). Parafoveal processing
within and between words. The Quarterly Journal of Experimental Psychology, 62, 1356 -
Just, M., & Carpenter, P. (1980). A theory of reading: from eye fixations to comprehension.
Psychological Review, 87, 329-354.
Kirby, J. A., Webster, L. A. D., Blythe, H. I., & Liversedge, S. P. (2008). Binocular
coordination during reading and non-reading tasks. Psychological Bulletin, 134, 742-763.
Kučera, H., & Francis, W. H. (1967). Computational analysis of present-day American English.
Providence, RI: Brown University Press.
Laing, E., Hulme, C., Grant, J., & Karmiloff-Smith, A. (2001). Learning to read in Williams
Syndrome: Looking beneath the surface of atypical reading development. Journal of Child
Psychology and Psychiatry, 42, 729-739.
Leonard, L. B. (1998). Children with Specific language Impairment. Cambridge: MIT Press.
Liu, Y., & Reichle, E. D. (2010). The emergence of adaptive eye movements in reading.
Proceedings of the Cognitive Science Society, (in press).
Liversedge, S. P., Rayner, K., White, S. J., Vergilino-Perez, D., Findlay, J. M., & Kentridge, R.
W. (2004). Eye movements when reading disappearing text: is there a gap effect in reading?
Vision Research, 44, 1013-1024.
McConkie, G. W., Kerr, P. W., Reddix, M. D., & Zola, D. (1988). Eye-movement control
during reading: I. The location of initial eye fixations on words. Vision Research, 28, 1107-
McConkie, G. W., Kerr, P. W., Reddix, M. D., Zola, D., & Jacobs, A. M. (1989). Eye
movement control during reading: II. Frequency of refixating a word. Perception &
Psychophysics, 46, 245-253.
McConkie, G. W., & Rayner, K. (1975). The span of effective stimulus during a fixation in
reading. Perception & Psychophysics, 17, 578-586.
McConkie, G. W., Zola, D., Grimes, J., Kerr, P. W., Bryant, N. R., & Wolff, P. M. (1991).
Children's eye movements during reading. In J. F. Stein (Ed.), Vision and visual dyslexia
(pp. 251-262). Boston: CRC Press.
Muter, V., Hulme, C., Snowling, M. J., & Stevenson, J. (2004). Phonemes, rimes, vocabulary
and grammatical skills as foundations of early reading development: Evidence from a
longitudinal study. Developmental Psychology, 40, 665-681.
Nation, K., & Snowling, M. J. (2004). Beyond phonological skills: broader language skills
contribute to the development of visual word recognition. Journal of Research in Reading,
Pollatsek, A., Reichle, E. D., & Rayner, K. (2006). Tests of the E-Z Reader model: Exploring
the interface between cognition and eye-movement control. Cognitive Psychology, 52, 1-56.
Rayner, K. (1975). The perceptual span and peripheral cues in reading. Cognitive Psychology,
Rayner, K. (1977). Visual attention in reading: Eye movements reflect cognitive processes.
Memory & Cognition, 4, 443-448.
Rayner, K. (1985). Do faulty eye movements cause dyslexia? Developmental Neuropsychology,
Rayner, K. (1986). Eye movements and the perceptual span in beginning and skilled readers.
Journal of Experimental Child Psychology, 41, 211-236.
Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research.
Psychological Bulletin, 124, 372-422.
Rayner, K. (2009). Eye movements and attention in reading, scene perception, and visual search.
The Quarterly Journal of Experimental Psychology, 62, 1457-1506.
Rayner, K., & Duffy, S. A. (1986). Lexical complexity and fixation times in reading: effects of
word frequency, verb complexity, and lexical ambiguity. Memory & Cognition, 14, 191-201.
Rayner, K., Inhoff, A. W., Morrison, R. E., Slowiaczek, M. L., & Bertera, J. H. (1981). Masking
of foveal and parafoveal vision during eye fixations in reading. Journal of Experimental
Psychology: Human Perception and Performance, 7, 167-179.
Rayner, K., Liversedge, S. P., & White, S. J. (2006). Eye movements when reading disappearing
text: The importance of the word to the right of fixation. Vision Research, 46, 310-323.
Rayner, K., Liversedge, S. P., White, S. J., & Vergilino-Perez, D. (2003). Reading disappearing
text: Cognitive control of eye movements. Psychological Science, 14, 385-388.
Rayner, K., McConkie, G. W., & Zola, D. (1980). Integrating information across eye
movements. Cognitive Psychology, 12, 206-226.
Rayner, K., & Raney, G. E. (1996). Eye movement control in reading and visual search: Effects
of word frequency. Psychonomic Bulletin & Review, 3, 245-248.
Rayner, K., Sereno, S. C., & Raney, G. E. (1996). Eye movement control in reading: A
comparison of two types of models. Journal of Experimental Psychology-Human Perception
and Performance, 22, 1188-1200.
Rayner, K., Well, A. D., Pollatsek, A., & Bertera, J. H. (1982). The availability of useful
information to the right of fixation in reading. Perception and Psychophysics, 31, 537-550.
Reichle, E. D., & Laurent, P. A. (2006). Using reinforcement learning to understand the
emergence of “intelligent” eye-movement behavior during reading. Psychological Review,
Reichle, E. D., Pollatsek, A., Fisher, D. L., & Rayner, K. (1998). Toward a model of eye
movement control in reading. Psychological Review, 105, 125-157.
Reichle, E. D., Pollatsek, A., & Rayner, K. (2006). E-Z Reader: A cognitive-control, serial-
attention model of eye-movement behavior during reading. Cognitive Systems Research, 7,
Reichle, E. D., Rayner, K., & Pollatsek, A. (1999). Eye movement control in reading:
accounting for initial fixation locations and refixations within the E-Z Reader model. Vision
Research, 39, 4403-4411.
Reichle, E. D., Rayner, K., & Pollatsek, A. (2003). The E-Z Reader model of eye-movement
control in reading: Comparisons to other models. Behavioral and Brain Sciences, 26, 445-
Reilly, R. G., & Radach, R. (2006). Some empirical tests of an interactive activation model of
eye movement control in reading. Cognitive Systems Research, 7, 34-55.
Salman, M. S., Sharpe, J. A., Eizenman, M., Lillakas, L., Westall, C., To, T., Dennis, M., &
Steinbach, M. J. (2006). Saccades in children. Vision Research, 46, 1432-1439.
Snowling, M.J. (2000). Dyslexia. Blackwell: Oxford.
Taylor, S. E. (1965). Eye movements while reading: Facts and fallacies. American Educational
Research Journal, 2, 187-202.
Taylor, S. E., Frackenpohl, H., & Pettee, J. L. (1960). Grade level norms for the components of
fundamental reading skill. EDL Research and Information Bulletin (Vol. 3). Huntington,
NY: Educational Development Laboratories.
Vitu, F., McConkie, G. W., Kerr, P., & O'Regan, J. K. (2001). Fixation location effects on
fixation durations during reading: an inverted optimal viewing position effect. Vision
Research, 41, 3513-3533.
Warren, T., White, S. J., & Reichle, E. D. (2009). Investigating the causes of wrap-up effects:
Evidence from eye movements and E-Z Reader. Cognition, 111, 132-137.
White, S.J., Rayner, K., & Liversedge, S.P. (2005). Eye movements and the modulation of
parafoveal processing by foveal processing difficulty: A reexamination. Psychonomic
Bulletin & Review, 12, 891-896.
Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia, and skilled
reading across languages: a psycholinguistic grain size theory. Psychological Bulletin, 131,
Table 1. Developmental changes in eye movements during reading. Measures marked * are
estimated from a graph. The data reported in Blythe et al. (2006, 2009) and Joseph et al. (2009)
are from children in the UK, where formal education begins at the age of 5-years. The data
reported in Blythe et al. (2010) and Häikiö et al. (2009) are from children in Finland, where
formal education begins at the age of 7-years. Finally, the data reported in Buswell (1922),
McConkie et al. (1991), Rayner (1986), and Taylor (1965) are from children in the US, where
formal education begins at the age of 6-years.