Morphological priming during reading: Evidence from eye movements
Kevin B. Paterson1
Alison Alcock1
&
Simon P. Liversedge2
1
2
School of Psychology, University of Southampton
Address correspondence to:
Kevin Paterson
School of Psychology,
University of Leicester,
University Road,
Leicester LE1 9HN, UK
Email: [email protected]
School of Psychology, University of Leicester
Morphological priming during reading
2
Abstract
We report an eye movement experiment that investigated whether prior exposure to
morphologically related and unrelated primes influenced processing of a target word that appeared
later in the same sentence. Prime-target pairs had a semantically transparent (e.g., marshy-marsh) or
only an apparent morphological relationship (e.g., secretary-secret), or were morphologically
unrelated but overlapped in orthography (e.g., extract-extra). Reading times for target words
revealed facilitation effects in measures of both early and late processing only for targets that
followed semantically transparent morphological primes, providing evidence of semantically
mediated priming between words read normally in a sentence. In addition, an increase in target word
skipping and in regressions from a post-target region when targets followed primes rather than
control words, regardless of the morphological relationship between the words, suggests that primetarget orthographic overlap influences parafoveal processing of target words. We discuss our
findings in relation to morphological priming during isolated word recognition and the process of
lexical identification during reading.
Key words: Eye movements during reading; word recognition; morphological decomposition;
morpho-orthographic processing, masked priming, unmasked priming.
Morphological priming during reading
3
It is widely argued that morphologically complex words are decomposed into their constituent
morphemes during visual word recognition, so that a word like marshy is recognised in terms of its
components parts (i.e., the stem word marsh and the grammatical morpheme -y; Feldman, 2000;
Feldman & Soltano, 1999; Longtin, Segui, & Hallé, 2003; Marslen-Wilson, Tyler, Waksler, & Older,
1994; Pastizzo & Feldman, 2002; Rastle, Davis, Marslen-Wilson, & Tyler, 2000; Rastle, Davis, &
New, 2004; Taft, 1981; Taft & Forster, 1975; Stanners, Neiser, Hernon, & Hall, 1979).
This observation has inspired considerable research into the effects of morphological structure
on word recognition, much of which has relied on priming methods based on the lexical decision
task to reveal the effects of presenting a morphologically complex word as a prime on the
recognition of a target word that is its constituent. This research has been conducted in the context of
other research showing that orthographic overlap between prime and target words can slow the
recognition of target words (e.g., Bijeljac-Babic, Biardeau, & Grainger, 1997; Colombo, 1986; Davis
& Lupker, 2006; Davis, Perea, & Acha, in press; Davis & Taft, 2005; De Moor & Brysbaert, 2000;
Drews & Zwiterlood, 1995; Grainger, Colé, & Segui, 1991; Grainger & Ferrand, 1994; Lupker &
Colombo, 1994; Segui & Grainger, 1990), where the slowdown in word recognition is usually
attributed to exposure to the prime providing activation to a lexical competitor for the target. By
contrast with these findings, research on morphological priming has shown that priming effects turn
facilitatory when prime and target words are morphologically related, so that presenting a
morphologically complex word such as marshy as a prime actually facilitates recognition of a stem
word target such as marsh. The reason the effect is facilitatory is thought to be due to decomposition
of the prime providing activation to a lexical entry for the root morpheme shared with the target word
rather than a lexical competitor, and so aiding rather than impeding word recognition (e.g., Feldman,
2000; Feldman & Soltano, 1999; Longtin et al., 2003; Longtin & Meunier, 2005; Marslen-Wilson et
al., 1994; Marslen-Wilson, Bozic, & Randall, 2008; Pastizzo & Feldman, 2002; Rastle et al., 2000;
Rastle et al., 2004; for a review, see Rastle & Davis, 2008).
Morphological priming during reading
4
An important issue in this research concerns whether morphological decomposition is mediated
by the semantic relationship between a word and its constituents, and an influential account of this
process holds that decomposition takes place only when the words are related in meaning (Giraudo &
Grainger, 2000; Marslen-Wilson et al., 1994; Plaut & Gonnerman, 2000; Rueckl & Raveh, 1999).
Accordingly, exposure to a word such as marshy should activate a representation of its stem (i.e.,
marsh), as this is a semantically related constituent, but exposure to secretary should not activate a
representation of secret, as secret and secretary are unrelated in meaning (in their modern usage).
Support for this account comes primarily from tasks, such as unmasked or cross-modal priming, in
which the prime is made available for conscious perception prior to the display of a target word (or
nonword). However, while the standard finding in this research is that facilitation is observed only
when prime-target pairs have a semantically transparent relationship (e.g., Longtin et al., 2003;
Marslen-Wilson et al., 1994; Rastle et al., 2000), other research employing a masked priming
procedure (e.g., Forster & Davis, 1984), in which a visual mask is shown first and the prime is
presented so briefly that it is unavailable for conscious report, has produced results that are consistent
with a process of morphological decomposition that is (at least initially) blind to word meaning
(Feldman, 2000; Feldman & Soltano, 1999; Longtin & Meunier, 2005; Marslen-Wilson et al., 2008;
Pastizzo & Feldman, 2002; Rastle & Davis, 2003; Rastle et al., 2000; Rastle et al., 2004; for a
review, see Rastle & Davis, 2008).
Rastle et al. (2004) reported an experiment that is typical of this masked priming research and
that showed that word recognition is facilitated when prime-target pairs have either a semantically
transparent morphological relationship (e.g., marshy-marsh) or only an apparent morphological
relationship (e.g., secretary-secret), but not when they are morphologically unrelated (e.g., extractextra). Various claims put forward in support of masked priming include that it avoids episodic
effects associated with unmasked priming techniques, and that effects in masked priming
experiments reflect early stages of word processing and so can reveal what properties of words are
Morphological priming during reading
5
available before they are consciously perceived (e.g., Forster, 1998). In line with these claims, Rastle
et al. (2004) argued that their findings, and those from similar studies, reveal that semantically
related and unrelated constituents both become activated during early stages of lexical access. It has
also been argued that such findings can be reconciled with unmasked priming research if it is
assumed that a meaning-independent decomposition procedure operates early in visual word
recognition and representations of inappropriate constituents are inhibited only later (Meunier &
Longtin, 2007; Rastle & Davis, 2008; Schreuder & Baayen, 1995).
It should be clear that research employing masked and unmasked priming provides important
insights into the influence of morphological priming on word recognition. However, a natural
extension of this research is to see if similar priming effects are observed in a reading situation in
which the prime and target words are read normally as part of a sentence. This was the approach
taken in the present research, which used measures of eye movements while reading to assess the
influence of prior exposure to a prime word presented earlier in a sentence on the processing of
morphologically related or unrelated target words that appeared later in the same sentence. One
motivation for this research was simply to see if effects similar to those reported in studies of isolated
word recognition occur in silent reading. However, a further reason for conducting this experiment
was to establish what influence priming has on eye movements during reading. There is considerable
evidence that eye movements are highly sensitive to processes involved in the lexical identification
of words, and that the decision about when to move the eyes is strongly associated with difficulty in
processing a fixated word (see Liversedge & Findlay, 2000; Rayner, 1998, 2009). Therefore, an
important concern for the present research was whether exposure to a prime word earlier in the
sentence would affect processes associated with the lexical identification of a target word and, in
turn, influence decisions about when to move the eyes. Such findings would make an important
contribution to our understanding of the influence of higher order language processing on the
identification of words during reading and help to inform the development of models of eye
Morphological priming during reading
6
movement control (e.g., Engbert, Nuthmann, Richter, & Kliegl, 2005; Reichle, Pollatsek, Fisher, &
Rayner, 1998; Reichle, Rayner, & Pollatsek, 2003; Reichle, Warren, & McConnell, 2009).
Previous eye movement research on morphological processing has focused on the effects of
morphological complexity on word processing (e.g., Juhasz, Starr, Inhoff, & Placke, 2003; Kambe,
2004; Lima, 1987) and the processing of compound words (e.g., Andrews, Miller, & Rayner, 2004;
Bertram & Hyönä, 2003; Hyönä & Pollatsek, 1998; Juhasz, 2007; Pollatsek, Hyönä, & Bertram,
2000), but has not investigated inter-word priming effects. Moreover, other eye movement research
that has examined inter-word priming during reading generally has been concerned with semantic or
identity/synonym priming (Binder & Morris, 1995; Camblin, Gordon, & Swaab, 2007; Carroll &
Slowiaczek, 1986; Morris, 1994; Morris & Folk, 1998; Traxler, Seely, Foss, Kaup, & Morris, 2000).
However, there is growing interest in whether lexical (and phonological) relationships between
words in sentences also give rise to priming effects and what influence this has on eye movements
(Carreiras, Ambrosio, & Meseguer, 2005; Frisson, Jamali, Pollatsek, & Meyer, 2009; Paterson,
Liversedge, & Davis, 2009; Williams, Perea, Pollatsek, & Rayner, 2006; Warren & Morris, 2009).
Paterson et al. (2009) reported an experiment typical of this research that provided particularly
strong evidence that inter-word lexical priming effects reported in word recognition research are also
observed in eye movements during reading. In this experiment, an orthographically related prime
word such as blur (or unrelated prime such as gasp) appeared just a few words earlier in a sentence
than a target word such as blue, in sentences such as “There was a blur/gasp as the blue lights of the
police car whizzed down the street”, and participants were instructed simply to read these sentences
as normal. The patterns of eye movements that were observed indicated that participants had more
difficulty in processing targets that followed related than unrelated primes. In particular, fixations
times were longer, and readers were more likely to make an inter-word regression back to earlier
sentence locations, when targets followed a related prime, and Paterson et al. took these findings to
show that inhibitory orthographic priming effects similar to those reported in studies of isolated word
Morphological priming during reading
7
recognition (e.g., Davis & Lupker, 2006; Segui & Grainger, 1990) also occur naturally due to
dependencies between words in sentences, and are reflected in eye movements during reading.
Paterson et al. (2009) argued that these findings are theoretically important, because they
demonstrate how the brief experience of processing one word (the prime) can rapidly influence the
lexical processing of a subsequent word (the target), which in turn affects decisions about when and
where to move the eyes. Moreover, the fact that the effect was observed in measures of eye
movements for words in sentences that were read normally indicates that eye movements are
sensitive to intra-sentential, inter-lexical influences that occur naturally during reading. This has
important implications for our understanding of how the process of lexical identification takes place
during reading, and is relevant to the ongoing debate about eye movement control during reading
(see, e.g., Reichle, Liversedge, Pollatsek, & Rayner, 2009), and whether decisions about when to
move the eyes are mediated by cognitive processing (e.g., an early stage of lexical access) or an
autonomous saccade generator (c.f., Engbert et al., 2005; Reichle et al., 1998, 2003). A critical
question in relation to these effects concerns the psychological mechanism by which these effects are
mediated, and the extent to which this mechanism is intrinsic to the system responsible for
oculomotor control during reading.
The present experiment employed a very similar procedure as Paterson et al. (2009) to
investigate morphological priming during reading, using prime, target, and control words from
stimuli used in the masked priming experiment reported by Rastle et al. (2004). As in their study,
prime-target pairs had either a semantically transparent morphological relationship (e.g., marshymarsh), an apparent morphological relationship (e.g., secretary-secret), or were morphologically
unrelated but overlapped in orthography (e.g., extract-extra). However, unlike in the study by Rastle
et al. (2004), the word pairs were embedded in sentence frames so that a prime (or control) word
appeared earlier in a sentence than a target word (see Table 1). These sentences were then shown to
participants, who were instructed to read them normally and we assessed the effects of the prime
Morphological priming during reading
8
manipulation on eye movement behaviour through the computation of reading time measures.
----------------------Table 1 about here-------------------While an important goal of our research was to replicate a key aspect of Paterson et al.’s
(2009) findings by firmly establishing that inter-word priming effects occur in silent reading, it was
also important to determine whether any priming effects were modulated by the morphological
relationship between prime and target words. This would provide further insights into the nature of
information that can carry across intervening words in a sentence to influence the lexical processing
of a subsequent word. In particular, the results would reveal whether lexical priming effects between
words in a sentence turn facilitatory when prime and target words are morphologically related, as
typically happens in word recognition studies (e.g. Rastle & Davis, 2008).
It was also of interest to determine whether priming effects are mediated by the semantic
relationship between prime and target words. In masked priming, prior exposure to a prime has been
shown to facilitate target word processing irrespective of whether prime and target words have a
semantically transparent or only an apparent morphological relationship (e.g., Rastle et al., 2004).
However, normal reading is very different to masked priming procedures, in that words that appear
early in a sentence are available for full inspection prior to the words downstream in the sentence
being read. Given this, it could be argued that normal reading more closely resembles procedures
used in unmasked priming, as prime words in this paradigm are made available for conscious
perception prior to the target word being processed. Consequently, it seemed likely that effects in the
present experiment would reflect circumstances in which readers have the opportunity to consciously
identify a prime before viewing the target word, and therefore produce a pattern of effects similar to
effects in unmasked priming studies. Therefore, as in unmasked priming research, target word
processing should be facilitated only when the morphological relationship between prime-target pairs
is semantically transparent. Finally, a major advantage of measuring eye movements during reading
is that this method provides a clear measure of the time course of effects. Thus, through the
Morphological priming during reading
9
computation of a range of eye movement measures we can assess when priming effects first emerge
in the eye movement record and how the influence of the prime changes over time and across
fixations. This analysis enables an assessment of when different sources of information, including
morphological prime information, first exert an influence on the processing of a word, and can shed
light on the nature and time course of the priming effects that occur between words during reading.
Method
Participants. Thirty-two native English speakers with normal uncorrected vision who were
undergraduates at the University of Leicester participated for course credits.
Design & Materials. Fifty-four sets of words were selected from the stimuli of Rastle et al.
(2004). These comprised 18 word pairs that had a semantically-transparent morphological
relationship, 18 pairs that had only an apparent morphological relationship, and 18 pairs that were
morphologically unrelated but overlapped in orthography. For each prime-target pair, the prime word
was matched for length and frequency with a control word that was unrelated to the target word.
Thus, two factors were manipulated: morphological priming condition (prime and target words had
either a semantically transparent morphological relationship, an apparent morphological relationship,
or were morphologically unrelated) and prime type (prime or control word).
Sentence frames were constructed so that prime-target pairs and control-target pairs could be
inserted into the same locations in each frame (see Figure 1). Before conducting the experiment, a
pre-test was conducted to assess the relative plausibility of sentence frames that contained either
prime-target or control-target pairs. Ten participants (who did not participate in the eye movement
experiment) rated the plausibility of these sentences on a 7-point Likert scale, where “1” indicated
that a sentence was entirely implausible and “7” indicated that it was entirely plausible. The results
clearly showed that sentences that contained either prime-target or control-target pairs did not differ
significantly in plausibility (6.5 vs. 6.4, F < 1). Moreover, the relatively high plausibility ratings
suggest that participants found the sentences to be acceptable English constructions. We also ensured
Morphological priming during reading
10
that the textual distance between prime / control words and target words was carefully controlled.
This was always identical for sentence frames containing either prime-target or control-target pairs,
and the textual distance (measured in alphabetic characters) between the prime / control words and
target words did not differ significantly across morphological priming conditions (semanticallytransparent morphological relationship = 14.7 characters, apparent morphological relationship = 14.2
characters, morphologically unrelated = 12.1 characters, F < 1.5). Finally, the sentences were
presented in counterbalanced lists so that each participant saw only one version of each sentence and
an equal number of sentences containing prime-target pairs (and control-target pairs) in each
morphological priming condition. Each list also contained 2 practice sentences that were presented at
the beginning of the experiment and an additional 80 sentences that were stimuli for an unrelated
experiment and were pseudo-randomly interleaved with sentences from the present experiment. A
full list of the experimental stimuli is included in the Appendix.
Apparatus. A Fourward Technologies Dual Purkinje Generation 6 Eye-tracker recorded
participants’ right eye movements during stimulus viewing. The eye-tracker has an angular
resolution of 10 min of arc and was interfaced with a PC that sampled fixation position every
millisecond. Sentences were presented as white text on a black background in Courier font on a 17
inch monitor. At the 80 cm viewing distance used in the experiment, three characters subtended
approximately 1 degree of visual angle. The sentences always started in the same location in the
upper left quadrant of the screen, and target words were always located close to the middle of a
sentence and so appeared near the centre of the screen.
Procedure. Before the start of the experiment, participants received an explanation of the
procedure. Participants were then seated at the eye-tracker and a bite-bar was used to prevent head
movements. A calibration procedure was then completed. Before the start of each trial, a fixation box
the same size as one alphabetic character appeared in the upper left quadrant of the screen. When
participants fixated this box, the experimenter initiated the presentation of a sentence, with the first
Morphological priming during reading
11
character of the sentence replacing the fixation box. The eye-tracker was re-calibrated if participants’
fixations did not match the fixation box. Participants were permitted to take breaks between trials.
On 25% of trials (including filler items), the presentation of a sentence was followed by a twoalternative forced-choice question that tested the participant’s comprehension of the sentence
content. Participants answered these questions by pressing the button corresponding to the correct
answer on a button box.
Results
Regions: Analyses were performed for three scoring regions (see Table 1): a pre-target region
comprising the word or phrase preceding the target word (which did not include the prime word), the
target word, and a post-target region that comprised the next word or next two words if the next word
was a short function word.
Analysis: Prior to the analysis of the eye movement data, an automatic procedure incorporated
fixations less than 80 ms into larger fixations within one character and deleted fixations less than 40
ms not within three characters of another fixation. Fixations over 1200 ms were truncated. In
addition, trials on which sentences were not read fully or that had tracker loss were eliminated by
deleting trials in which no first-pass fixations were made in adjacent scoring regions (accounting for
4.2% of trials). Accuracy for the comprehension questions was high, above 90%, indicating that
participants had read and fully understood the sentences. No items were excluded from the eye
movement analyses based on participants’ answers to the comprehension questions. A range of
standard eye movement measures was then computed for each scoring region (see Rayner, 1998,
2009). For each region, we report first fixation durations (duration of the first fixation on a word),
gaze durations (summed duration of all fixations from first fixating a word before a saccade from it referred to as first-pass reading times for regions containing multiple words), and total reading times
(the summed duration of all fixations in a region). In addition, single fixation durations (the duration
of the fixation on a word receiving only one first-pass fixation), word skipping (the frequency of
Morphological priming during reading
12
trials in which a word was not fixated) and regressions in (the frequency of regressions into a region)
are reported for target words, and the frequency of regressions out of a region are reported for target
words and for the post-target region. Analyses of first-pass reading time measures for each scoring
region excluded trials in which no fixations were made within that region.
Data for each scoring region were subjected to a 3 (morphological priming condition:
semantically transparent morphological relationship, apparent morphological relationship,
morphologically unrelated) x 2 (prime type: prime, control) repeated measures ANOVA, computing
error variance over participants (F1) and sentences (F2). Sentences in different morphological
priming conditions were composed of different words, including different prime and target words.
Consequently, any main effects of the morphological priming condition could reflect theoretically
uninteresting differences in the processing of these words. Therefore, we will report only main
effects of prime type and interaction effects, since these provide the clearest indication of priming
effects in each morphological priming condition. Also, in order to present the priming effects most
transparently, a mean priming effect (PE) magnitude is included alongside the mean eye movement
data for each morphological priming condition (see Table 2).
----------------------Table 2 about here---------------------Pre-target region: There were no significant priming effects in this region, Fs < 1.
Target word: There was a main effect of prime type in first fixation durations, F1(1,31) = 6.79,
p < .05, p2 = .18, and F2(1,51) = 4.97, p < .05, p2 = .09, and single fixation durations, F1(1,31) =
6.79, p < .05, p2 = .22, and F2(1,51) = 8.21, p < .01, p2 = .14, that interacted with the
morphological relationship between prime-target pairs (first fixation durations: F1(2,62) = 5.93, p <
.01, p2 = .33, and F2(2,51) = 3.75, p < .05, p2 = .13; single fixation durations: F1(2,62) = 4.25, p <
.05, p2 = .27, and F2(2,62) = 3.16, p = .05,p2 = .10). Tukey tests revealed significant facilitatory
priming only when prime and target words were semantically related, p < .001. Although the main
effect of prime type was marginal in gaze durations for target words, F1(1,31) = 3.50, p = .07,p2 =
Morphological priming during reading
13
.10, and F2(1,51) = 3.80, p = .06, p2 = .07, and not reliable in total reading times for target words,
Fs < 1.6, there were significant interactions in both measures (gaze durations: F1(2,62) = 3.21, p <
.05, p2 = .22, and F2(2,51) = 3.33, p < .05, p2 = .12; total reading times: F1(2,62) = 4.23, p < .05,
p2 = .22, and F2(2,62) = 3.95, p < .05, p2 = .13). Again, Tukey tests revealed significant facilitatory
priming only when prime-target pairs were semantically related, p < .01. Thus, early measures of
reading time (i.e., first fixation durations, gaze durations) for target words revealed a semantically
mediated priming effect, and this effect was also observed in total reading times for these words.
There was a main effect of prime type in target word skipping, F1(1,31) = 5.55, p < .05, p2 =
.15, and F2(1,51) = 4.08, p < .05, p2 = .07, that did not interact with the morphological relationship
between prime-target pairs, Fs < 1. Note that participants will have made the decision to skip target
words prior to these words being fixated; that is, based on information about the target word
extracted from the parafovea. The word skipping effect showed that target words that followed
primes rather than control words were skipped more frequently (regardless of the semantic relation
between the prime and target). Finally, although the regression data revealed an increase in
regressions to target words that followed primes rather than control words, this effect was not
significant, Fs < 2.4.1 No other priming effects were significant at this region, Fs < 2.
Post-target region: There was a main effect of prime type in regressions from the post-target
region, F1(1,32) = 9.31, p < .01, p2 = .23, and F2(1,51) = 9.47, p < .01,p2 = .18, that did not
interact with the morphological relationship between prime and target words, Fs < 1. More
regressions were made from the post-target region when target words followed primes than control
words. Other research has reported an increase in regressions following word skipping (Drieghe,
Brysbaert, Desmet, & De Baecke, 2004) and, in line with this observation, further analyses showed
that more regressions were made from the post-target region when target words were skipped (38%)
than when target words received a first-pass fixation (24%; t1(32) = 3.30, p < .01, and t2(53) = 2.14,
Morphological priming during reading
14
p < .01). However, it should be clear that word skipping cannot fully account for the regression effect
at the post-target region, as an analysis of regressions from this region when target words were not
skipped also produced a significant main effect of prime type, F1(1,32) = 8.68, p < .01, p2 = .22,
and F2(1,51) = 7.27, p < .01, that did not interact with the morphological relationship between prime
and target words, Fs < 1. Thus, it appears that the progression of the eyes through the sentence was
disrupted on a proportion of trials, even when the target words was fixated, indicating that
orthographic overlap between prime and target words disrupted target word identification, and that
this disruption did not occur exclusively in cases where the target word was skipped.
The main effect of prime type in first-pass reading times for the post-target region was not
significant, Fs < 1.7. However, a marginal interaction in first-pass reading times, F1(2,62) = 2.92, p
= .06, p2 = .18, and F2(2,51) = 3.52, p < .05, p2 = .12, and a marginal main effect of prime type in
total reading times, F1(1,31) = 3.60, p = .07, p2 = .10, and F2(1,51) = 3.73, p = .06, p2 = .06. The
effect in first-pass reading times was of particular interest, and post-hoc Tukey tests confirmed that it
was driven primarily by post-target difficulty when target words followed morphologically unrelated
primes rather than control words (p = .06), and no significant priming effects were observed in the
other morphological priming conditions (ps > 12). Thus it appears that orthographic similarity
between prime and target words disrupted processing of the post-target region, but only when primetarget pairs were morphologically unrelated. No other priming effects were significant, Fs < 1.
Discussion
Perhaps the most immediately striking feature of our data was the priming effect in fixation
times for target words. This revealed that fixation times were substantially shorter for targets that
followed a morphologically related prime than a control word, but only when prime-target pairs were
related in meaning. The experiment therefore provides very clear evidence for a semantically
mediated priming effect in silent reading. Moreover, as this effect emerged in eye movement
measures that reflect early word processing (i.e., first fixation durations, single fixation durations,
Morphological priming during reading
15
and gaze durations for target words), it seems clear that prior exposure to a morphologically complex
word in a sentence can fairly immediately facilitate processing of a semantically related stem word
that appears a few words later in the same sentence.
This aspect of our findings accords well with other experiments showing that lexical priming
effects reported in studies of word recognition occur naturally, and perhaps frequently, between
words in sentences that are read normally (e.g., Carreiras et al., 2005; Carroll & Slowiaczek, 1986;
Paterson et al., 2009), and that the brief experience of processing a prime word can carry across
intervening words to influence the processing of a target word that appears later in the same
sentence. Moreover, this inter-word priming has a very rapid influence on the lexical identification
process, and this in turn affects decisions about when to move the eyes. Most importantly, the current
data provide further compelling evidence that eye movements during reading are sensitive to intrasentential, inter-lexical priming influences. Such findings are highly relevant to the on-going debate
about eye movement control during reading (e.g., Reichle et al., 2009) and indicative of the centrality
of lexical identification to decisions regarding saccade initiation during reading.
The pattern of results in fixation times for the target words is complementary to findings from
research on isolated word recognition. Where this research has used masked priming, it has shown
that morphological priming effects are not mediated by word meaning (e.g., Rastle et al., 2004).
However, other research in which prime words are presented without a mask at durations that allow
them to be consciously identified prior to a target word being displayed has shown that word
recognition is facilitated only when prime-target pairs have a semantically transparent morphological
relationship (e.g., Longtin et al., 2003; Marslen-Wilson et al., 1994; Rastle et al., 2000). That the
pattern of fixation times in our experiment most closely resembles the findings from unmasked
priming is unsurprising as, in both paradigms, the prime words are fully visible, and are therefore
available for conscious identification prior to fixation on the target word. It is argued that conscious
identification of the prime in unmasked priming research results in inappropriate constituents being
Morphological priming during reading
16
inhibited prior to a response being made for the target word (e.g., Meunier & Longtin, 2007; Rastle
& Davis, 2008; Schreuder & Baayen, 1995). Thus, as readers in our experiment were able to
consciously identify prime words before encountering the target, it seems likely that this resulted in
representations of inappropriate prime word constituents being inhibited prior to fixation on the
target word, and this would explain why facilitatory priming effects were observed in fixation times
for target words in the present experiment only when the relationship between the prime and target
words was semantically transparent. However, it has been important in morphological priming
research to separate the influence of the morphological, orthographic, and semantic relationships
between prime and target words (see, e.g., Marslen-Wilson et al., 1994; Rastle et al., 2000),
particularly as words in English that have a derivational morphological relationship are also
orthographically and semantically related. This is also a concern in relation to the present
experiment. Although it is clear from the present results that the morphological priming effect we
observed in fixation times for target words is not driven by the orthographic relationship between
prime and target words, further research is needed to disentangle whether this effect derives from the
morphological or semantic relationship between these words or both.
The priming effect in fixation times for target words was clear and makes sense in the context
of previous research on isolated word recognition. However, effects that emerged in other eye
movement measures provided an unexpected insight into other aspects of inter-word priming
influences on eye movements during reading. In particular, there was very clear evidence of a
priming effect both in skipping rates for target words and in regressions from the post-target region.
These showed that target words that followed a prime rather than a control word were less likely to
be fixated during first-pass processing and, in addition, that readers were more likely to make a
regressive saccade from the post-target region back to earlier sentence locations when targets
followed primes rather than control words. As these priming effects were not mediated by the
morphological relationship between prime-target pairs, and emerged primarily in target word
Morphological priming during reading
17
skipping and regressions from the post-target region (and only to some extent in first-pass reading
times for this region), it seems likely that they represent an influence that is independent from the
semantically mediated morphological priming observed in fixation times for target words. Instead,
the effects may have been due solely to the orthographic relationship that existed between prime and
target words in all three priming conditions. To be clear, it was fundamental to the logic of our
experiment that prime-target pairs (but not control-target pairs) overlapped in orthography, not only
when these words had a semantically transparent relationship or an apparent morphological
relationship, but also when they were unrelated either semantically or apparently morphologically
(e.g., extract & extra). Given that a priming effect occurred in all three conditions, it appears that this
orthographic overlap, rather than any higher order relationship between the prime and target words,
was responsible for the skipping and regression effects. Moreover, as these effects emerged in eye
movement behaviours that are typically prevalent during text reading, and are not present in standard
isolated word recognition experiments, they may reveal aspects of lexical processing that are unique
to normal text reading.
Word skipping, in particular, provides an insight into the processing of words in parafoveal
vision. Considerable evidence suggests that readers extract information about the identities of words
to the right of fixation in parafoveal vision, and that this information is used to plan the next saccade
and to begin processing of the next words in a sentence (Balota, Pollatsek, & Rayner, 1985; Binder,
Pollatsek, & Rayner, 1999; Rayner, 1975; Juhasz, White, Liversedge, & Rayner, 2008). Moreover,
research has shown that readers typically do not fixate all of the words in a sentence and it is widely
argued that words that are highly predictable and easy to identify in parafoveal vision do not have to
be fixated (e.g., Binder et al., 1999, Balota et al., 1985; Drieghe, Brysbaert, Desmet, & Debaecke,
2004; Dreighe, Rayner, & Pollatsek, 2005; Rayner, Binder, Ashby & Pollatsek, 2001; Rayner &
Well, 1996; White, Rayner, & Liversedge, 2005; for a review, see Brysbaert, Drieghe, & Vitu,
2005). The present findings add to this research by revealing that words are also more likely to be
Morphological priming during reading
18
skipped if they closely follow another word that overlaps substantially in orthography. Indeed, it
seems that the brief experience of processing a word just a few words earlier in a sentence can carry
over intervening words to influence parafoveal processing of a subsequent word, and to the extent
that words are more likely to be identified (or possibly misidentified) in parafoveal vision, they are
less likely to be initially fixated.
Research on word skipping has also revealed that when a word is skipped this often is quickly
followed by a regression back to earlier sentence locations and researchers have attributed this
pattern of behaviour to the initiation of an error-correcting procedure (Drieghe et al., 2004; see also
Brysbaert et al., 2005). Consistent with this observation, we found that readers made more inter-word
regressions from the post-target region when target words were skipped compared to when they were
fixated. Thus, it seems likely that word-skipping in our experiment might also have led to the
initiation of an error-correcting procedure. Indeed, and as we have already suggested, target words
that follow an orthographically similar prime may sometimes be misidentified in parafoveal vision,
and the subsequent detection of this misperception may increase the likelihood of readers making a
regression to re-process preceding text. However, this argument is qualified by the finding that
readers also made more regressions from the post-target region when targets followed prime than
control words, even when the target word was not skipped. Thus, it should be clear that the word
skipping effect does not fully account for word misperception effects that occurred during reading,
and that orthographic overlap between prime and target words appears to have disrupted target word
identification even in cases where target words were fixated.
The effect observed in first-pass reading times for the post-target region may add to our
understanding of the influence of orthographic overlap on target word perception. This effect showed
that reading times were inflated when targets followed primes only when prime-target pairs were
morphologically unrelated, and so appears to reveal that orthographic overlap was most harmful
when prime and target words were morphologically unrelated. One possibility is that the
Morphological priming during reading
19
morphological relationship between prime-target words in the other conditions helped overcome
misidentification of the target words by facilitating access to the target word’s representation
following the initial misidentification, but that readers had a harder time recovering from
misidentification when the words were morphologically unrelated because access to target word
representations was not facilitated in this case.
That intra-sentential, inter-word orthographic priming causes words to be misidentified in
parafoveal vision may not be surprising, and is consistent with growing evidence that words are
misperceived in reading quite frequently (e.g., Perea & Pollatsek, 1998; Pollatsek, Perea, & Binder,
1999; Slattery, 2010). Much of this evidence comes from eye movement research that has assessed
the influence of a word’s orthographic neighbourhood on word processing, where a word’s
orthographic neighbours are usually defined as words that differ by just one letter while preserving
letter order and length (e.g., Coltheart, Davelaar, Jonasson, & Besner, 1977; but for an expanded
definition, see Davis et al., 2009; Davis & Taft, 2005). The approach taken in this research has been
to examine eye movements for sentences that contain a target word such as spice that has a higher
frequency neighbour (e.g., space), and has shown that these words receive longer fixations and more
fixations after a regression than control words that do not have a higher frequency neighbour (Perea
& Pollatsek, 1998; Pollatsek, Perea, & Binder, 1999; Slattery, 2010). These effects are usually
explained in terms of readers initially mistaking the target word for its higher frequency neighbour,
and thus causing disruption to subsequent processing of the sentence. One possibility is that exposure
to an orthographically similar word just a few words earlier in the sentence might also cause readers
to sometimes misperceive a similar upcoming word in parafoveal vision, and this misperception
might disrupt sentence processing. Such an account could explain the findings from our experiment.
Note, however, that the degree of orthographic overlap between prime-target words, and the fact that
this overlap was always at the beginnings of words, may have been fundamental to eliciting the
effects we observed. Indeed, it is widely argued that beginning letter information is particularly
Morphological priming during reading
20
important in parafoveal pre-processing of words (e.g., Briihl & Inhoff, 1995; Rayner, McConkie, &
Zola, 1980; Rayner, Well, Pollatsek, & Bertera, 1982; White & Liversedge, 2005), but that
parafoveal pre-processing is insensitive to semantic or morphological word information (Altarriba,
Kambe, Pollatsek, & Rayner, 2001; Bertram & Hyönä, 2007; Juhasz et al., 2008; Kambe, 2004;
Lima, 1987; Rayner, Balota, & Pollatsek, 1986). These findings are directly relevant to the current
results. It seems likely that the considerable overlap in orthography at the beginnings of words in all
three priming conditions drove the word skipping effect, and consequently, this effect held for
stimuli in all conditions regardless of the semantic or morphological relationships that may also have
existed between primes and targets (although the disruption this effect caused may have been less
severe when there was a morphological relationship between prime and target words). Nevertheless,
the contribution of orthographic priming to word skipping remains to be fully determined,
particularly since skipping rates have not always been reported in studies examining inter-word
lexical priming during reading (e.g., Camblin et al., 2007; Carreiras et al., 2005; Paterson et al, 2009;
Traxler et al., 2000). Further research clearly is needed to establish the full range of circumstances in
which word skipping occurs and how pre-processing of words in parafoveal vision contributes to the
phenomenon.
Another issue for future research concerns the source of inter-word lexical priming effects.
Paterson et al. (2009) argued that effects in their experiment could be explained by two different
mechanisms. The first was that the effects were due to ongoing patterns of activity in the reader’s
mental lexicon, consistent with the account of priming effects in competitive network models of
word recognition (e.g., Davis, 1999, 2003; Grainger & Jacobs, 1996, McClelland & Rumelhart,
1981). A fundamental assumption of these models is that activation of a lexical entry decays over
time and is inhibited when other lexical entries become activated as other words are read. Therefore,
these models can explain priming effects that carry over intervening words so long as the activation
of a prime word’s lexical entry has not decayed or been inhibited substantially in the interval
Morphological priming during reading
21
between identification of the prime and fixation on the target word. According to this account, the
fact that priming effects were observed in the present experiment indicates that there was sufficient
ongoing activation of a prime word’s lexical entry for this to affect the activation of the related, but
different, target word such that identification was influenced. Furthermore, this pattern of activation
must have been maintained throughout the period between processing of the two words (i.e., when
the prime and target words were separated by only small number of completely independent words).
In the present experiment the prime and target words were on average 2.6 words apart. However, our
priming effects may also be explained by episodic memory accounts which propose that when prime
and target words are read separately (as in the present experiment), the processing of one word (the
prime) instantiates an episodic memory trace which subsequently impacts on identification during
processing of a similar word (the target) read downstream in the sentence. In this way, the memory
trace that is evoked at the target influences current word processing (and for a review of the episodic
approach to lexical priming, see Tenpenny, 1995). Clearly, a comprehensive account of inter-word
priming during reading must be able to discriminate between these alternative accounts and research
underway in our, and probably other, laboratories will attempt to do this. Indeed, there is also
considerable interest in establishing the degree to which priming effects can carry across intervening
words in word recognition experiments (Forster, 2008).
In sum, the present research has produced novel evidence relating to morphological and
orthographic priming effects in eye movements during reading. This research has revealed some
similarities with findings from studies of isolated word recognition. However, where the present
results differ from isolated word recognition findings, they reveal an influence of inter-word priming
on aspects of lexical processing that are unique to reading text and cannot manifest in isolated word
recognition research. In this way, our findings extend current understanding and provide insight into
the process of lexical identification as it occurs normally during natural reading.
Morphological priming during reading
22
Footnote
1. Further analyses were conducted to determine if regressions were made back to the prime word.
There were negligible differences in the frequency of regressions in prime and control word
conditions in either semantically transparent (19.9% vs. 19.6%), apparent (28.1% vs. 24.7%), or
morphologically related conditions (25.4% vs. 26.7%), and neither the main effect of prime type nor
the interaction of prime type and morphological relationship was statistically reliable (Fs < 1).
Morphological priming during reading
23
References
Altarriba, J., Kambe, G., Pollatsek, A., & Rayner, K. (2001). Semantic codes are not used in
integrating information across eye fixations in reading: Evidence from fluent Spanish-English
bilinguals. Perception & Psychophysics, 63, 875–890.
Andrews, S., Miller, B., & Rayner, K. (2004). Eye movements and morphological segmentation of
compound words: There is a mouse in mousetrap. European Journal of Cognitive Psychology,
16, 285-311.
Balota, D. A., Pollatsek, A., & Rayner, K. (1985). The interaction of contextual constraints and
parafoveal visual information in reading. Cognitive Psychology, 17, 364-390.
Bertram, R., & Hyönä, J. (2003). The length of a complex word modifies the role of morphological
structure: Evidence from eye movements when reading short and long Finnish compounds.
Journal of Memory and Language, 48, 615-634.
Bertram, R. & Hyönä, J. (2007). The interplay between parafoveal preview and morphological
processing in reading. In R. van Gompel, M. Fischer, W. Murray & R.Hill (eds.), Eyemovements: A window on mind and brain (pp. 391-407). Amsterdam: Elsevier.
Bijeljac-Babic, R., Biardeau, A., & Grainger, J. (1997). Masked orthographic priming in bilingual
word recognition. Memory & Cognition, 25, 447–457.
Binder, K. S., & Morris, R. K. (1995). Eye movements and lexical ambiguity resolution: Effects of
prior encounter and discourse topic. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 21, 1-11.
Binder, K. S., Pollatsek, A., & Rayner, K. (1999). Extraction of information to the left of the fixated
word in reading. Journal of Experimental Psychology: Human Perception and Performance,
25, 1162–1172.
Briihl, D., & Inhoff, A. W. (1995). Integrating information across fixations during reading: The use
of orthographic bodies and of exterior letters. Journal of Experimental Psychology: Learning,
Morphological priming during reading
24
Memory, and Cognition, 21, 55–67.
Brysbaert, M., Drieghe, D., & Vitu, F. (2005). Word skipping: Implications for theories of eye
movement control in reading. In G. Underwood (Ed.), Cognitive processes in eye guidance (pp.
53-77). Oxford: Oxford University Press.
Camblin, C. C., Gordon, P. C., & Swaab, T. Y. (2007). The interplay of discourse congruence and
lexical association during sentence processing: Evidence from ERPs and eye tracking. Journal
of Memory and Language, 56, 103-128.
Carreiras, M., Perdomo, A., & Meseguer, E. (2005). Are stem homographs and orthographic
neighbours processed differently during silent reading? Language and Cognitive Processes, 20,
317-339.
Carroll, P., & Slowiaczek, M. L. (1986). Constraints of semantic priming in reading: A fixation time
analysis. Memory & Cognition, 14, 509-522.
Colombo, L. (1986) Activation and inhibition with orthographically similar words. Journal of
Experimental Psychology: Human Perception and Performance, 12, 226–234.
Coltheart, M., Davelaar, E., Jonasson, J. F., & Besner, D. (1977). Access to the internal lexicon. In S.
Dornic (Ed.), Attention and performance VI (pp. 535–555). Hillsdale, NJ: Erlbaum.
Davis, C. J. (1999). The self-organising lexical acquisition and recognition (SOLAR) model of visual
word recognition. Unpublished doctoral dissertation, University of New South Wales.
Davis, C. J. (2003). Factors underlying masked priming effects in competitive network models of
visual word recognition. In S. Kinoshita & S. J. Lupker (Eds.), Masked Priming: The State of
the Art. Philadelphia: Psychology Press.
Davis, C. J., & Lupker, S. J. (2006). Masked inhibitory priming in English: Evidence for lexical
inhibition. Journal of Experimental Psychology: Human Perception and Performance, 32,
668–687.
Davis, C. J., Perea, M., & Acha, J. (2009). Re(de)fining the orthographic neighbourhood: The role of
Morphological priming during reading
25
addition and deletion neighbours in lexical decision and reading. Journal of Experimental
Psychology: Human Perception & Performance, 35, 1550-1570.
Davis, C. J., & Taft, M. (2005). More words in the neighborhood: Interference in lexical decision due
to deletion neighbors. Psychonomic Bulletin & Review, 12, 904-910.
De Moor, W., & Brysbeart, M. (2000). Neighborhood-frequency effects when primes and targets are
of different lengths. Psychological Research/Psychologische Forschung, 63, 159–162.
Drews, E., & Zwitserlood, P. (1995). Morphological and orthographic similarity in visual word
recognition. Journal of Experimental Psychology: Human Perception and Performance, 21,
1098–1116.
Drieghe, D., Brysbaert, M., Desmet, T., & Debaecke, C. (2004). Word skipping in reading: On the
interplay of linguistic and visual factors. European Journal of Cognitive Psychology, 16, 79103.
Drieghe, D., Rayner, K., & Pollatsek, A. (2005). Eye movements and word skipping during reading
revisited. Journal of Experimental Psychology: Human Perception and Performance, 31, 954–
969.
Engbert, R., Nuthmann, A., Richter, E., & Kliegl, R. (2005). SWIFT: A dynamical model of saccade
generation during reading. Psychological Review, 112, 777-813.
Feldman, L. B. (2000). Are morphological effects distinguishable from the effects of shared meaning
and shared form? Journal of Experimental Psychology: Learning, Memory, and Cognition, 26,
1431-1444.
Feldman, L. B., & Soltano, E. (1999). Morphological priming: The role of prime duration, semantic
transparency, and affix position. Brain and Language, 68, 33-39.
Forster, K. I. (1998). The pros and cons of masked priming. Journal of Psycholinguistic Research,
27, 203–233.
Forster, K. I. (2008). The intervenor effect in masked priming: How does masked priming survive
Morphological priming during reading
26
across an intervening word? Journal of Memory and Language, 60, 36-49.
Forster, K. I., & Davis, C. (1984). Repetition priming and frequency attenuation in lexical access.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 10, 680-698.
Frisson. S., Jamali, L., Pollatsek, A., & Meyer, A. (2009). Reading Dr. Seuss: Rhyme helps, then
hurts, processing. Poster presented at the 15th European Conference on Eye Movements, 23rd27th August, Southampton, UK.
Giraudo, H., & Grainger, J. (2000). Effects of prime word frequency and cumulative root frequency
in masked morphological priming. Language and Cognitive Processes, 15, 421-444.
Grainger, J., & Ferrand, L. (1994). Phonology and orthography in visual word recognition: Effects of
masked homophone primes. Journal of Memory and Language, 33, 218–233.
Grainger, J., Colé, P., & Segui, J. (1991). Masked morphological priming in visual word recognition.
Journal of Memory and Language, 30, 370–384.
Hyönä, J., & Pollatsek, A. (1998). Reading Finnish compound words: Eye fixations are affected by
component morphemes. Journal of Experimental Psychology: Human Perception and
Performance, 24, 1612-1627.
Juhasz, B. J. (2007). The influence of semantic transparency on eye movements during English
compound word recognition. In R. P. G. van Gompel, M. H. Fischer, W. S. Murray, & R. L.
Hill (Eds.), Eye movements: A window on mind and brain (pp. 373-389). Amsterdam: Elsevier.
Juhasz, B. J., Starr, M. S., Inhoff, A. W., & Placke, L. (2003). The effects of morphology on the
processing of compound words: Evidence from naming, lexical decisions and eye fixations.
British Journal of Psychology, 94, 223-244.
Juhasz, B. J., White, S. J., Liversedge, S. P., & Rayner, K (2008). Eye movements and the use of
parafoveal word length information in reading. Journal of Experimental Psychology: Human
Perception and Performance, 34, 1560–1579.
Kambe, G. (2004). Parafoveal processing of prefixed words during eye fixations in reading: Evidence
Morphological priming during reading
27
against morphological influences on parafoveal preprocessing. Perception & Psychophysics,
66, 279–292.
Lima, S. D. (1987). Morphological analysis in sentence reading. Journal of Memory and Language,
26, 84–99.
Liversedge, S. P. & Findlay, J. M. (2000). Eye movements reflect cognitive processes. Trends in
Cognitive Sciences, 4, 6-14.
Longtin, C. –M., & Meunier, F. (2005). Morphological decomposition in early visual word
processing, Journal of Memory and Language, 53, 26–41.
Longtin, C. –M., Segui, J., & Hallé, P. A. (2003). Morphological priming without morphological
relationship. Language and Cognitive Processes, 18, 313-334.
Lupker, S. J., & Columbo, L. (1994). Inhibitory effects in form priming - Evaluating a phonological
competition explanation. . Journal of Experimental Psychology: Human Perception and
Performance, 20, 437-451.
Marslen-Wilson, W. D., Bozic, M., & Randall, B. (2008). Early decomposition in visual word
recognition: Dissociating morphology, form, and meaning. Language and Cognitive Processes,
23, 394-421.
Marslen-Wilson, W. D., Tyler, L. K., Waksler, R., & Older, L. (1994). Morphology and meaning in
the English mental lexicon. Psychological Review, 101, 3-13.
McClelland, J. L., & Rumelhart, D. E. (1981). An interactive activation model of context effects in
letter perception: Pt. 1. An account of basic findings. Psychological Review, 88, 375–407.
Meunier, F., & Longtin, C. M. (2007). Morphological decomposition and semantic integration in
word processing. Journal of Memory and Language, 56, 457-471.
Morris, R. K. (1994). Lexical and message level sentence context effects on fixation times in reading.
Journal of Experimental Psychology: Learning, Memory and Cognition, 20, 92-103.
Morris, R. K. & Folk, J. R. (1998). Focus as a contextual priming mechanism. Memory & Cognition,
Morphological priming during reading
28
26, 1313-1322.
Pastizzo, M., & Feldman, L. B. (2002). Discrepancies between orthographic and unrelated baselines
in masked priming undermine a decompositional account of morphological facilitation. Journal
of Experimental Psychology: Learning, Memory, and Cognition, 28, 244-249.
Paterson, K. B., Liversedge, S. P., & Davis, C. J. (2009). Inhibitory neighbor priming effects in eye
movements during reading. Psychonomic Bulletin & Review, 16, 43-50.
Perea, M., & Pollatsek, A. (1998). The effects of neighborhood frequency in reading and lexical
decision. Journal of Experimental Psychology: Human Perception and Performance, 24, 767–
779.
Plaut, D. C., & Gonnerman, L. M. (2000). Are non-semantic morphological effects incompatible with
a distributed connectionist approach to lexical processing? Language and Cognitive Processes,
15, 445-485.
Pollatsek, A., Hyönä, J., & Bertram, R. (2000). The role of morphological constituents in reading
Finnish compound words. Journal of Experimental Psychology: Human Perception and
Performance, 26, 820-833.
Pollatsek, A., Perea, M., & Binder, K. (1999). The effects of neighborhood size in reading and lexical
decision. Journal of Experimental Psychology: Human Perception and Performance, 25, 1142–
1158.
Rastle, K., & Davis, M. H. (2008). Morphological decomposition based on the analysis of
orthography. Language and Cognitive Processes, 23, 942-971.
Rastle, K., Davis, M. H., & New, B. (2004). The broth in my brother's brothel: Morpho-orthographic
segmentation in visual word recognition. Psychonomic Bulletin & Review, 11, 1090-1098.
Rastle, K., Davis, M. H., Marslen-Wilson, W. D., & Tyler, L. K. (2000). Morphological and semantic
effects in visual word recognition: A time course study. Language and Cognitive Processes,
15, 507-538.
Morphological priming during reading
29
Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research.
Psychological Bulletin, 124, 372-422.
Rayner, K. (2009). The Thirty-fifth Sir Frederick Bartlett Lecture: Eye movements and attention in
reading, scene perception, and visual search. Quarterly Journal of Experimental Psychology,
62, 1457-1506.
Rayner, K., Balota, D. A., & Pollatsek, A. (1986). Against parafoveal semantic preprocessing during
eye fixations in reading. Canadian Journal of Psychology, 40, 473–483.
Rayner, K., Binder, K. S., Ashby, J., & Pollatsek, A. (2001). Eye movement control in reading: word
predictability has little influence on initial landing positions in words. Vision Research, 41,
943-954.
Rayner K,, McConkie, G. W., & Zola, D. (1980). Integrating information across eye fixations.
Cognitive Psychology. 12, 206–226.
Rayner, K., & Well, A. D. (1996). Effects of contextual constraint on eye movements during reading:
a further examination. Psychonomic Bulletin & Review, 3, 504-509.
Rayner, K., Well, A. D., Pollatsek, A., & Bertera, J. H. (1982). The availability of useful information
to the right of fixation in reading. Perception & Psychophysics, 31, 537–550.
Reichle, E. D., Liversedge, S. P., Rayner, K., & Pollatsek, A. (2009). Encoding multiple words
simultaneously in reading is implausible. Trends in Cognitive Sciences, 13,115-119..
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., 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, 445-476.
Reichle, E. D., Warren, T., & McConnell, K. (2009). Using E-Z Reader to model the effects of higerlevel language processing on eye movements during reading. Psychonomic Bulletin & Review,
16, 1-21.
Morphological priming during reading
30
Rueckl, J. G., & Raveh, M. (1999). The influence of morphological regularities on the dynamics of a
connectionist network. Brain and Language, 68, 110-117.
Schreuder, R., & Baayen, R. H. (1995). Modelling morphological processing. In L. B. Feldman (Ed.),
Morphological aspects of language processing (pp. 131-154). Hillsdale, NJ: Erlbaum.
Segui, J., & Grainger, J. (1990). Priming word recognition with orthographic neighbors: Effects of
relative prime–target frequency. Journal of Experimental Psychology: Human Perception and
Performance, 16, 65–76.
Slattery, T. (2010). Word misperception the neighbor frequency effect and the role of sentence
context: An eye movement investigation. Journal of Experimental Psychology: Human
Perception and Performance, 35, 1969-1975.
Stanners, R. F., Neiser, J. J., Hernon, W. P., & Hall, R. (1979). Memory representation for
morphologically related words. Journal of Verbal Learning and Verbal Behavior, 18, 399-412.
Taft, M. (1981). Prefix stripping revisited. Journal of Verbal Learning and Verbal Behavior, 20, 289297.
Taft, M., & Forster, K. I. (1975). Lexical storage and retrieval of prefixed words. Journal of Verbal
Learning and Verbal Behavior, 14, 638-647.
Tenpenny, P. L. (1995). Abstractionist versus episodic theories of repetition priming and word
identification. Psychonomic Bulletin & Review, 2, 339-363.
Traxler, M. J., Seely, R. E., Foss, D. J., Kaup, B., & Morris, R. K. (2000). Priming in Sentence
Processing: Intralexical Spreading Activation, Schemas, and Situation Models. Journal of
Psycholinguistic Research, 29,581-595.
Warren, S., & Morris, R. K. (2009). Phonological similarity effects in reading. Paper presented at the
15th European Conference on Eye Movements, 23rd-27th August, Southampton, UK.
White, S. J., & Liversedge, S. P. (2004). Orthographic familiarity influences initial eye fixation
positions in reading. European Journal of Cognitive Psychology, 16, 52-78.
Morphological priming during reading
31
White, S.J., Rayner, K., & Liversedge, S.P. (2005). The influence of parafoveal word length and
contextual constraint on fixation durations and word skipping in reading. Psychonomic Bulletin
& Review, 12, 466-471.
Williams, C. C., Perea, M., Pollatsek, A., & Rayner, K. (2006). Previewing the neighborhood: the
role of orthographic neighbors as parafoveal previews in reading. Journal of Experimental
Psychology: Human Perception and Performance, 32, 1072–1082.
Morphological priming during reading
Author Notes
We are grateful to Raymond Bertram, Keith Rayner, and an anonymous reviewer for helpful
comments on an earlier draft of this paper.
32
Morphological priming during reading
33
Appendix
Sentences are shown with prime and control words in brackets and target words underlined but were
displayed normally in experiment.
Semantically transparent morphological relationship
1. The Cabinet said in a written statement that the tax was due to the current (government /
situation) but they could not govern unless people were willing to pay.
2. The sculpture in the park fascinated one (viewer / ranger) but spoilt the view of the dale.
3. Because she broke her arm, the (trainee / cookery) teacher could not train until the beginning of
the next academic term.
4. The forest had a (marshy / thorny) path leading to a marsh where students studied wildlife.
5. There was only one (mourner / tripper) to mourn the final sailing of a famous cruise ship.
6. On the sofa was a cream (tufted / silken) throw with a tuft of red in each corner.
7. To avoid appearing (inhibitory / amateurish), he would always inhibit his actions and opinions.
8. Next door to John lived a (widowed / beastly) lady, who became a widow when her husband died
in an accident.
9. The teachers saw advantages to the (agreement / equipment), but couldn’t agree on who would be
responsible for it.
10. At the end of her road trip, the (northern / friendly) girl headed up north back to her home.
11. The man bought a (creamy / watery) tea and some cream scones at the cafe.
12. Because the defendant looked so (guilty / formal), his guilt seemed obvious to everyone in the
court.
13. The critics thought that the (artistry / calmness) of the artist was evident in the waves in the
painting.
14. When we arrived at the (chilly / finely) restored barn the chill kept us awake despite warm
clothes and hot drinks.
15. Looking at the cottage, she said, “It may be (greenery / snobbish), but I hate green around the
door.”
16. The new (teacher / trainee) finally started to teach her group after they settled down.
17. Although the company had a new (employer / addition), they would still employ the current
workers.
18. Although they expected a harsh (reaction / physical), to react the way the woman did was entirely
unreasonable.
Apparent morphological relationship
19. It was true that the company (secretary / obviously) believed the secret and didn’t know what to
do.
20. The search for the perfect wedding cake became (fruitless / alcoholic), as fruit cake made with
brandy was all that was available.
21. The sound of the (trumpet / chatter) would trump the singer’s poor performance.
22. The policeman investigated the old (crooked / pottery) shop, but the crook had not taken
anything.
23. He left the house with the (courteous / developer) apprentice then went to court to collect the
legal documents.
24. The garden was lovely, apart from the (thicket / scruffy) bush that was thick and overgrown.
25. Jane was told to resign from the music (department / production) and to depart without notice.
26. Depending on the budget and (boarder / factual) information, full board may be an option.
27. The electrical (flicker / advisor) warned that a flick of the switch could be fatal.
28. Only by (rational / steadily) thinking can the ration of supplies be distributed fairly amongst the
Morphological priming during reading
34
needy people.
29. If the man had kept the key (number / really) safe he would not be numb with sorrow now.
30. It was argued that setting an essay (question / actually) aids the quest to find the best applicant
for a management position.
31. You can buy (brandy / safely) from a range of brand new suppliers through secure web sites.
32. The vendor said it would make an excellent (brisket / foundry) ensuring brisk business with the
customer.
33. Hazardous situations are not outlined by the (treaty / angler), so ensure you treat them with
caution.
34. The hockey player performed well and was (plucky / winger) today, so to pluck her from the
team would be a mistake.
35. During the search, the police found the (coaster / muffler) on the far coast of the island.
36. The girl wanted a toy, so asked the (trolley / naughty) boy to get a troll off the shelf.
Morphologically unrelated
37. The man complained about the (rabbit / weekly) hunt to the Rabbi in the local village.
38. The lady wore a dress made of a (plush / filmy) material, plus a shawl to keep her warm.
39. Jane thought the boy looked very (weird / manly) standing by the weir in the stream.
40. She left the dance (studio / gently) holding the stud earring that she had found.
41. I made the Christmas decorations (twinkle / cheaply) for my twin sisters and their friends.
42. Although it was a (stubborn / moisture) stain, the stub of soap removed it.
43. The meeting discussed the (international / revolutionary) plan, but the intern didn’t understand.
44. When he examined the (glade / cuffs), he was glad to see there was no evidence of the scuffle
that had taken place.
45. The girl sat in the corner was (sight / happy) reading and gave a sigh when she turned.
46. Jo kept quiet about her involvement in the (brothel / warfare) and ate her broth in silence.
47. They were stopped in their tracks by the (squawk / oddity), but the squaw who emerged from the
wigwam was not to be feared.
48. A boy pulled at the brown (jerkin / twisty) fabric with a jerk to get the man’s attention.
49. It took some (shovel / tricky) work to shove the beehive from the tree without upsetting the bees.
50. The junior doctor was told about the (candidacy / epileptic) and was candid in the interview.
51. More time was allowed to (extract / justify) the extra information that was needed.
52. The researcher conducted a (phonetic / dreadful) analysis by phone, rather than in person.
53. The boy felt pain in his foot at the (surface / medical) as the surf had caused him serious injury.
54. Despite the order being (heaven / firmly) sent, he couldn’t heave the sword from the rock.
Morphological priming during reading
35
Table 1: Example stimuli and region divisions
Sentences are shown with prime and control words in brackets and target words underlined but were
displayed normally in experiment. Vertical lines denote region boundaries.
Semantically transparent morphological relationship
The forest had a (marshy / thorny) path| leading to a| marsh| where| students studied wildlife.
Apparent morphological relationship
It was true that the company (secretary / obviously)| believed the| secret| and didn’t| know what to do.
Morphologically unrelated
More time was allowed to (extract / justify)| the| extra| information| that was needed.
36
Morphological priming during reading
Table 2: Mean eye movement measures for each condition (PE indicates size and direction of priming effect*)
Morphological Relationship of Prime and Target Words
Semantically-Transparent
Apparent
Unrelated
Prime
Control
PE
Prime
Control
PE
Prime
Control
PE
First fixation durations
264
259
-5
261
261
0
268
260
-8
First pass reading times
345
352
+7
315
315
0
327
316
-11
Total reading times
427
432
+5
445
430
-15
442
449
+7
First fixation duration
243
280
+37
267
265
-2
263
278
+15
Single fixation duration
245
286
+41
267
270
+3
266
285
+19
Gaze durations
272
308
+36
296
295
-1
294
304
+10
Total reading times
327
374
+47
416
400
-16
411
421
+10
Word skipping (%)
20.0
14.2
+5.8
16.8
13.1
+3.7
21.8
18.6
+3.2
Regressions in (%)
15.3
13.2
-2.1
26.2
22.8
-3.4
27.9
24.1
-3.8
Regressions out (%)
7.5
12.7
+5.2
15.0
14.6
-0.4
16.9
14.3
-2.6
Pre-Target Region
Target Region
37
Morphological priming during reading
Post-Target Region
First fixation durations
251
257
+6
257
264
+7
260
255
-5
First pass reading times
434
448
+14
436
448
+12
437
405
-32
Total reading times
557
546
-11
608
597
-11
604
547
-57
Regressions out (%)
17.8
13.8
-4.0
36.7
28.2
-8.5
33.5
26.0
-7.5
* Shorter reading times, greater word-skipping rates, and lower regression rates in prime versus control conditions are labelled as positive values and
vice versa.