=Paper=
{{Paper
|id=Vol-1347/paper42
|storemode=property
|title=The role of grammar factors and visual context in Norwegian children's pronoun resolution
|pdfUrl=https://ceur-ws.org/Vol-1347/paper42.pdf
|volume=Vol-1347
|dblpUrl=https://dblp.org/rec/conf/networds/FoynVE15
}}
==The role of grammar factors and visual context in Norwegian children's pronoun resolution==
https://ceur-ws.org/Vol-1347/paper42.pdf
The role of grammar factors and visual context in Norwegian
children’s pronoun resolution
Camilla Hellum Foyn Mila Vulchanova Rik Eshuis
Department of Department of Department of
language and literature language and literature language and literature
NTNU NTNU NTNU
camilla.foyn mila.vulchanova hendrik.eshuis
@ntnu.no @ntnu.no @ntnu.no
1 Introduction
Example of the stimulus sentences:
Most personal pronouns have one entry in the
mental lexicon, but they can have different refer-
1. Introduction sentence:
ents depending on the context they appear in.
They are sometimes fairly ambiguous. There is
Der er hesten og kaninen
also evidence that pronoun resolution is impaired
There are the.horse and the.rabbit
in many developmental deficits. Children have to
learn how to find the intended referent, but we do 2a. Subject-cleft:
not know much about how resolution strategies
are acquired. How do visual context and syntac- Det er hesten som kiler kaninen
tic context influence children’s pronoun pro- It is the.horse that tickles the.rabbit
cessing? Using eye-tracking, we investigate for
the first time the development of Norwegian 2b. Object-cleft:
children’s pronoun resolution competencies in
their L1. Det er kaninen hesten kiler
It is the.rabbit the.horse tickles
2 The study
The participants were monolingual 3-, 5-, and 7- 3. Ambiguous pronoun sentence:
year-old children, as well as a control group of
monolingual adults. There were between 25 and Han kan telle til ti
28 participants in each group. In the first of three He can count to ten
experiments, they listened to it-cleft sentences
4. Question sentence:
with either subject focus (2a) or object focus
(2b), while they watched illustrations of two an-
Hvem kan telle til ti?
imals (corresponding to the subject and the ob-
Who can count to ten?
ject) on a screen. It-clefts provide a good envi-
ronment for testing syntactically expressed focus,
and appear to be more frequent in Norwegian
Conditions
than e.g., English (Gundel, 2002). The animals
were sometimes shown performing the actions 1 Subject-cleft Depicted action
from the cleft-sentences, and other times not (see
2 Subject-cleft No depicted action
Table 1 for overview of conditions). Thereafter,
the participants heard an ambiguous pronoun
3 Object-cleft Depicted action
sentence (3), and eye-tracking data were collect-
ed to determine whether they looked at the sub-
3 Object-cleft No depicted action
ject or object referent. In addition, offline data
were collected, by asking the participants to
Table 1: Conditions.
name or point at the pronoun referent (4).
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
187
3 Earlier findings et al. (2014) found, namely that young children
show a first-mention bias that is too slow to de-
According to Järvikivi et al. (2013), German 4- tect, or it may simply show that 3-year-olds are
year-olds and adults show a subject preference too young to comprehend cleft-sentences. In any
regardless of which word the it-cleft focuses on. case, this shows that older children have a
Moreover, children seem to show a weaker sub- stronger preference for the focused referent than
ject preference than adults. We expect similar younger children do.
results from our data.
Adults showed an overall subject preference re-
Hartshorne et al. (2014) discovered that 2- to 3- gardless of sentence type, except in the condition
year-olds have a first-mention preference that with object-cleft and no depicted action. This
seldom is detected because they take longer to appears to be the only condition that weakens
process. We thus expect young children to show their subject preference, probably because it
a preference for subject and/or first-mentioned leaves the subject without syntactic focus and
character, albeit at a later time window, whereas with no visual support. Thus, the effect of syn-
adults will show an earlier preference than chil- tactic focus and/or a first-mention preference
dren. emerges here.
Bittner and Kuehnast (2011) have found that Moreover, depicted action seems to have dis-
German 3-year-olds rely more on context-cues tracted the adults, since the effect of subject vs.
than older German children, who more often use object-clefts offline was only found when the
syntax-cues. We thus expect that young children action was not depicted.
will be more influenced by the presence of visual
context, whereas older children will be more sen- In subject-clefts as opposed to object-clefts, 5-
sitive to syntactically expressed focus. and 7-year-olds displayed an online subject pref-
erence, although in different manners. Adults
4 Results also showed this preference, both offline and
online. Hence, all these three age groups appear
A mixed design ANOVA showed that 5-year- to use syntax cues, but adults seem to be more
olds looked more at the subject referent after aware of them, as 5- and 7-year-olds still only
subject-clefts than object-clefts from 500-1000 reveal their preferences through their gaze be-
ms after pronoun onset (p > .05), whereas adults havior. This supports Järvikivi et al.’s (2013)
did the same during the first 500 ms (p = .06). suggestion that children use the same cues as
Adults also showed a general subject preference adults, but that they have not fully developed
both offline (p > .001) and online (p > .05), spe- their ability to do so.
cifically after subject-clefts as opposed to object-
clefts offline (p > .05). Moreover, first-look data References
(first look at subject or object referent after pro-
noun onset) revealed a stronger subject prefer- Dagmar Bittner and Milena Kuehnast. 2011. Compre-
hension of intersentential pronouns in child German
ence in 7-year-olds after subject-clefts than ob-
and child Bulgarian. First Language, 32(1-2), 176–
ject-clefts (p > .05). We found no significant ef- 204.
fect of visual context in the children. However,
an interaction effect in adults showed that their Jeanette K. Gundel. 2002. Information structure and
stronger subject preference in subject-clefts than the use of cleft sentences in English and Norwegian.
object-clefts offline was only present when the Language and Computers, 39(1), 113–128.
action was not depicted (p > .05).
Joshua K. Hartshorne, Rebecca Nappa, & Jesse
5 Conclusions Snedeker. 2014. Development of the first-mention
bias. Journal of Child Language, 41(3), 1-24.
The results from the time series data suggest that
adults process the pronouns faster than children, Juhani Järvikivi, Pirita Pyykkönen-Klauck, Sarah
which supports Hartshorne et al. (2014). Schimke, Saveria Colonna, & Barbara Hemforth.
2013. Information structure cues for 4-year-olds and
In contrast to the older children, the 3-year-olds adults: tracking eye movements to visually presented
performed at chance level in all the different anaphoric referents. Language and Cognitive Pro-
cesses, 0(0), 1–16.
conditions. This may be due to what Hartshorne
188