=Paper=
{{Paper
|id=Vol-1347/paper30
|storemode=property
|title=Language proficiency moderates morphological priming in children and adults
|pdfUrl=https://ceur-ws.org/Vol-1347/paper30.pdf
|volume=Vol-1347
|dblpUrl=https://dblp.org/rec/conf/networds/HasenackerBS15
}}
==Language proficiency moderates morphological priming in children and adults==
https://ceur-ws.org/Vol-1347/paper30.pdf
Language proficiency moderates morphological priming in children
and adults
Jana Hasenäcker Elisabeth Beyersmann Sascha Schroeder
Max-Planck-Institute Laboratoire de Psychologie Max-Planck-Institute
for Human Development, Cognitive, Aix-Marseille for Human Development,
Berlin Université and Centre de la Berlin
hasenaecker@mpib- Recherche Scientifique, sascha.schroeder@mpib-
berlin.mpg.de Marseille berlin.mpg.de
lisi.beyersmann
@gmail.com
Only few studies have been concerned with
1 Introduction morphological decomposition in beginning read-
ers. The few studies from English and French
A number of studies have shown that skilled
used complex word primes, pseudocomplex
readers decompose morphologically complex
word primes and non-morphological word
words upon encountering them (for a review, see
primes. Quémart, Casalis and Colé (2011) found
Rastle & Davis, 2008). It has been proposed that
priming in French grade 3, 5, and 7 children
this segmentation process is early and automatic
from complex as well as pseudocomplex words,
and is driven by orthographic form, while being
thus suggesting that children already use adult-
blind to semantic content, thus also called mor-
like decomposition processes. In contrast, Bey-
pho-orthographic (Rastle, Davis, & New, 2004;
ersmann, Castles and Coltheart (2012) only
Taft, 2003). One key finding in favor of this
found priming from truly complex words in
proposition comes from masked morphological
grade 3 and 5 English-speaking children, indicat-
priming: the recognition of a target word is fa-
ing that morpho-orthographic priming is not au-
cilitated when it is preceded by a morphological-
tomatized yet and decomposition relies more on
ly related word prime (teacher-TEACH). Facili-
semantics in developing readers. However, no
tation has also been found in a number of lan-
studies with children have used complex
guages for targets preceded by pseudocomplex
pseudoword primes so far, although they provide
word primes that is words that appear to have a
the possibility to utilize the paradigm in lan-
morphologically complex structure, but are sim-
guages that do not naturally have pseudocomplex
plex words (corner-CORN). Moreover, facilita-
words, such as German.
tion has as well been observed from complex
Morphological decomposition in German
pseudoword primes, that is a non-existing com-
can be insightful to investigate, because of its
bination of a stem and affix (flexify-FLEX). For
language specific characteristics. German has a
non-morphological nonword primes, that is a
transparent orthography and is morphologically
non-existing combination of a word and a non-
rich. As a consequence, morphological entities
morphemic ending (flexint-FLEX), mixed results
might present a very useful unit for effective
have been obtained (Longtin & Meunier, 2005;
word recognition, even for beginning readers.
Morris, Porter, Grainger & Holcomb, 2011). Re-
Nevertheless, for children being still in the pro-
cent evidence from French points to a moderat-
cess of reading acquisition and showing more
ing role of language proficiency: the magnitude
variability in their lexical representations, lan-
to which morpho-orthographic information is
guage proficiency can be expected to play an
used increases as a function of individual vocab-
even greater role than Beyersmann et al. (2014)
ulary and spelling skills in adults (Andrews &
found for adults.
Lo, 2013; Beyersmann, Casalis, Ziegler &
The aim of the present study was there-
Grainger, 2014).
fore to test whether the moderating effect of lan-
Copyright © by the paper’s authors. Copying permitted for private and academic purposes.
In Vito Pirrelli, Claudia Marzi, Marcello Ferro (eds.): Word Structure and Word Usage. Proceedings of the NetWordS Final
Conference, Pisa, March 30-April 1, 2015, published at http://ceur-ws.org
132
guage proficiency, as indexed by vocabulary and matched on length. For each target all four types
spelling skills, on morphological priming can be of primes were created. Primes were matched on
replicated with German adults and whether it length, suffix length and non-morphemic ending
generalizes to readers at the lower end of the pro- length across conditions. Four counterbalanced
ficiency range, namely children. We expect to lists of prime-target combinations were created,
see evidence for a more automatized form of each containing a target only once, such that par-
morpho-orthographic decomposition in highly ticipants saw each target only with one of the
proficient children (replicating the Quémart et al. four prime conditions.
pattern), whereas low-skilled children should
show less priming (as in Beyersmann et al., 2.3 Procedure
2012) or no robust priming at all. In our adult Stimuli were presented in white 20-point Courier
group, we expect robust priming in all three New font in the center of a black screen on a 15″
prime conditions (including the nonsuffixed con- laptop monitor with a refresh rate of 60 Hz. Each
dition) in high proficiency participants, but re- trial consisted of a 500-ms fixation cross, fol-
duced non-suffixed priming in low proficiency lowed by a 500-ms forward mask of hash keys,
participants then a prime in lowercase for 50 ms, followed by
the target in uppercase. The target remained on
screen until response. Participants were instruct-
2 Method ed to indicate whether the presented stimuli was
2.1 Participants an existing German word or not by pressing a
key as quickly and as accurately as possible.
Twenty-four university students (13 women, Mage They were not informed about the presentation
= 25.2 years, age range: 20–29 years) and 24 el- of the prime.
ementary school children (13 girls, Mage = 9.5
years, age range: 8;6–10;9 years, grade 3-5) par- 2.4 Results
ticipated in the experiment. Reaction times were analyzed using linear
Each participant’s language proficiency mixed-effects modeling. Participants and items
was assessed, using a spelling and a vocabulary were included as random factors and lexical sta-
test. Adults performed a spelling recognition test, tus of the target (word, pseudoword), prime type
which was modelled after the one used by An- (suffixed word, suffixed pseudoword, nonsuf-
drews and Lo (2012). Participants were asked to fixed pseudoword, unrelated word), age group
classify 100 words as correctly or incorrectly (adults, children) and language proficiency (con-
spelled. Children performed the fill-in-the-gap tinuous measure combined of the spelling and
dictation test of the SLRT-II (Moll & Landerl, vocabulary scores), as well as all their interac-
2010). For assessment of vocabulary, adults tions, were included as fixed effects. Where ap-
completed the German version of the LexTALE propriate, one-sided post-hoc contrasts were ap-
(Lemhöfer & Broersma, 2012), and children the plied comparing all related priming conditions
vocabulary subtest of the CFT 20 (Weiß, 1998). with the unrelated condition. For contrasting
A composite measure of spelling and vocabulary readers with higher and lower proficiency, reac-
was calculated by standardizing and averaging tion times of participants scoring one standard
the spelling and vocabulary scores for each par- deviation above or below the mean proficiency
ticipant. measure within their age group were used. Sig-
2.2 Materials nificance was evaluated using the normal distri-
bution. Results are reported for word targets on-
We conducted a masked priming lexical decision ly. Descriptive statistics are provided in Table 1.
experiment using real suffixed words (kleidchen- For adults, priming was observed from all
KLEID), suffixed pseudowords (kleidtum- three related conditions (suffixed word, suffixed
KLEID), nonsuffixed pseudowords (kleidekt- pseudoword and nonsuffixed pseudoword) rela-
KLEID) and unrelated controls (träumerei- tive to the unrelated condition, z=5.04, z=4.43,
KLEID) as primes. 50 word targets were selected z=2.07, all p<.05. However, language proficien-
from the childLex corpus (Schroeder, Würzner, cy moderated priming effects. Priming in the
Heister, Geyken, & Kliegl, 2014) and 50 nonsuffixed pseudoword condition was only sig-
pseudoword targets were created by changing nificant for adults with higher language profi-
one letter from a real word that was not in the ciency (+1SD), z=1.74, p<.05, but not for adults
target word set. Word and nonword targets were
133
RT
Prime Type Adults Children Stimulus Example
All participants
Suffixed Word 593 (12) 1024 (36) kleidchen - KLEID
Suffixed Nonword 597 (12) 1051 (38) kleidtum - KLEID
Nonsuffixed Nonword 614 (13) 1045 (38) kleidekt - KLEID
Unrelated 629 (14) 1087 (41) träumerei - KLEID
Higher Language Proficiency (+1SD)
Suffixed Word 588 (12) 900 (28) kleidchen - KLEID
Suffixed Nonword 583 (12) 924 (30) kleidtum - KLEID
Nonsuffixed Nonword 602 (12) 903 (28) kleidekt - KLEID
Unrelated 620 (13) 974 (33) träumerei - KLEID
Lower Language Proficiency (-1SD)
Suffixed Word 599 (12) 1189 (48) kleidchen - KLEID
Suffixed Nonword 613 (13) 1218 (51) kleidtum - KLEID
Nonsuffixed Nonword 626 (14) 1239 (52) kleidekt - KLEID
Unrelated 638 (14) 1229 (51) träumerei - KLEID
Table 1. Response times (in ms) for children and adults,
averaged across items for each participant. Standard errors are presented in parentheses.
with lower language proficiency (-1SD), z=1.16, Moreover, the pattern of priming generalizes to
p=.25. beginning readers with higher language profi-
For children, proficiency played an even ciency: they show priming similar to that of
more pronounced role than for adults: higher higher proficient adults. For children with lower
proficiency children (+1SD) showed the same language proficiency, the effects did not reach
pattern as higher proficiency adults, namely significance, but were clearly most pronounced
priming from all related condition, z=3.03, in the suffixed word condition.
z=2.02, z=2.96, all p<.05. In contrast, in lower We argue that there is a developmental
proficiency children (-1SD) priming in none of gradient in the use of morphological information
the conditions reached significance, although during reading acquisition, driven by language
there was a numerical advantage from suffixed proficiency. Beginning readers with low lan-
word primes in the mean reaction times (40ms guage proficiency are only able to benefit from
faster compared to the unrelated condition). morpho-semantic information, if at all. More
advanced lexical knowledge allows readers to
3 Conclusion extract morpho-orthographic information. Fol-
lowing Andrews and Davis (1999) and Grainger
Our results confirm recent evidence for French
and Ziegler (2011), we assume that this happens
adults (Beyersmann et al., 2014), showing that
through segmentation of the affix in lower profi-
the extent to which morphological information is
ciency adults, as indicated by the priming effects
exploited depends on language proficiency also
of both suffixed prime conditions. Crucially,
in German. Adults in the present study showed
higher proficiency adult and even child readers
morphological priming effects from suffixed
with sophisticated lexical knowledge are able to
word primes (kleidchen-KLEID), suffixed
additionally use segmentation of the embedded
pseudoword primes (kleidtum-KLEID) and also
stem, therefore showing facilitation also in the
nonsuffixed pseudoword primes (kleidekt-
nonsuffixed prime condition. Our results high-
KLEID) relative to unrelated words (träumerei-
light the importance of lexical knowledge as a
KLEID). Priming from the nonsuffixed
further determinant of the ability to exploit mor-
pseudoword condition did not continue to be sig-
nificant with decreasing language proficiency.
134
phological structure in the word recognition pro- orthographic segmentation in visual word recogni-
cess, especially in children. tion. Psychonomic Bulletin & Review, 11, 1090-
1098.
Reference
Schroeder, S., Würzner, K.-M., Heister, J., Geyken,
Andrews, S., & Davis, M. H. (1999). Interactive acti- A., & Kliegl, R. (2014). childLex: A lexical data-
vation accounts of morphological decomposition: base for German read by children. Manuscript
Finding the trap in mousetrap? Brain and Lan- submitted for publication. Max Planck Institute for
guage, 68, 355-361. Human Development, Berlin.
Andrews, S., & Lo, S. (2013). Is morphological prim-
Taft, M. (2003). Morphological representation as a
ing stronger for transparent than opaque words? It
correlation between from and meaning. In E. As-
depends on individual differences in spelling and
sink & D. Sandra (Eds.) Reading complex words
vocabulary. Journal of Memory and Language, 68,
(pp. 113-137). Amsterdam, The Netherlands:
279-296.
Kluwer.
Beyersmann, E., Casalis, S., Ziegler, J. C., & Grain-
ger, J. (2014). Language proficiency and morpho- Weiß, R. H. (1998). Grundintelligenztest Skala 2
orthographic segmentation. Psychonomic Bulletin (CFT 20) mit Wortschatztest (WS) und Zahlenfol-
Review. gentest (ZF). Göttingen: Hogrefe.
Beyersmann, E., Castles, A., & Coltheart, M. (2012).
Morphological processing during visual word
recognition in developing readers: Evidence from
masked priming. The Quarterly Journal of Exper-
imental Psychology, 65, 1306-1326.
Grainger, J., & Ziegler, J. C. (2011). A dual-route
approach to orthographic processing. Frontiers in
Psychology, 45,
Lemhöfer, K., & Broersma, M. (2012). Introducing
LexTALE: A quick and valid Lexical Test for Ad-
vanced Learners of English. Behavior Research
Methods, 44, 325-343.
Longtin, C. M., & Meunier, F. (2015). Morphological
decomposition in early visual word processing.
Journal of Memory and Language, 53, 26-41.
Moll, K., & Landerl, K. (2010). SLRT-II: Lese- und
Rechtschreibtest: Weiterentwicklung der Salzbur-
ger Lese- und Rechtschreibtests (SLRT): Manual.
Bern: Verlag Hans Huber.
Morris, J., Porter, J. H., Grainger, J., & Holcomb, P. J.
(2011). Effects of lexical status and morphological
complexity in masked priming: An ERP study.
Language & Cognitive Processes, 26, 558-599.
Quémart, P., Casalis, S., & Colé, P. (2011). The role
of form and meaning in the processing of written
morphology: A priming study in French develop-
ing readers. Journal of Experimental Child Psy-
chology, 109, 478-496.
Rastle, K., & Davis, M. H. (2008). Morphological
decomposition based on the analysis of orthogra-
phy. Language & Cognitive Processes, 23, 942-
971.
Rastle, K., Davis, M. H., & New, B. (2004). The
broth in my brother’s brothel: Morpho-
135