Transcript Document

Perceiving and encoding motion events
in children of English and Mandarin
Shanju Lin
Amanda J. Owen Van Horne
Dept of Communication Sciences and Disorders & DeLTA Center
The University of Iowa
Introduction
Motion events
• One object moves with respect to another object.
Component
Manner
Path
Talmy, 1975, 1985
• Cognition and language in motion events
are
early
developed.
run, walk, jump, fly, roll
- 18 months olds detect object,
go, come, leave, exit,
trajectory, and the whole motion.
Example
through, over, under
Ground:
The boy jumped from
•Source & Goal the tree /to the ground.
Raskin & Poulin-Dubois, 2002
- Language-specific expressions in
spatial language early
Choi & Bowerman, 1991; Berman & Slobin, 1994
• Learning a language influences cognitive development by providing a
representational resource.
Clark, 2004
Cognition and Language in Motion Events
• Salience Hypothesis: Learning a language permanently changes cognitive
attention to spatial distinctions.
Bowerman & Choi, 2003, Slobin, 2004, Clark, 2004
• Underspecification Hypothesis: Cognition is only transiently affected by
language specific patterns; Once it is nonlinguistic in nature performance
is not affected by language specific information
Gennari et al., 2002, Papafragou et al., 2002, Papafragou & Selimis, 2010
General Methods
Nonlinguistic Categorization Task:
Which of two alternative videos goes with the baseline video
24 triads of videos: Each triad has one baseline and
two alternative variants.
Differ minimally on Manner, Path, Source, or Goal.
Manners: Minimal pairs constructed within category
Nonvehicle: jump, walk, roll
Vehicle: drive car, ride motorcycle, fly airplane
Paths: through, under, over
Sources & Goals: 2 of each present in each
baseline. Minimal pairs switched locations, not items.
Baseline
Manner Change
Goal Change
Linguistic Description Task:
•
•
•
•
Describe the 24 baseline videos
Coding: 1) How many elements did the speaker mention?
2) Which elements did the speaker mention?
He went from the ferris wheel under the rollercoaster to the drinks – 3 items S, P, G
The lizard drives under the tracks – 2 items, M & P
He was driving – 1 item, M
The lizard went to a bottle – 1 item, G
Tasks adapted from Papafragou et al. (2002) & Gennari et al. (2002)
Cross-linguistic Comparison: English & Mandarin
Linguistic Typology
Talmy (1991), Slobin (2004)
1. Verb-framed languages: Spanish, Italian, Greek
La mina ENTRO a la casa CORRIENDO (Spanish) ‘The girl entered the house running’
2. Satellite-framed languages: English, German
The girl RAN INTO the house
3. Equipollently-framed languages: Mandarin, Thai
Nuhei PAO JIN CHU fangzi li. ‘Girl run enter go house inside’. (Mandarin)
- Mandarin-speaking children rely mainly on this serial verb
construction when encoding motion events.
Lin, 2006
Q1: Does linguistic encoding affect cognitive perception in complex motion
events?
Q2: Do Mandarin speakers encode motion events differently from English
speakers when those events are less likely to be described in the
Mandarin canonical construction?
Participants
English
N
Age
PPVT
KBIT/TONI
TD4
19
4;8
(4;0 – 4;10)
111.95
(97 - 131)
104.68
(92 - 130)
Mandarin
Adults
19
29;2
(23;4 – 57;10)
TD4
14
4;6
(4;3 – 4;11)
112.21
(102 - 129)
113.43
(100 – 137)
-
Adults
13
29;7
(26;3 – 34;6)
-
Results
Categorization
No other differences
apparent
Numer of times
Children tend to notice
source more than adults in
both languages
12
10
8
M-TD4
E-TD4
M-Adult
E-Adult
6
4
2
Grand: X2 (9, 65) = 20.66, p< .001,
S<M=P=G
Age: X2 (3, 65) = 16.74, p< .001,
Source: Adults < TD4
Language: X2 (3, 65) = 2.41, p> .05
0
Manner
Bias
Path Bias Goal Bias
Source
Bias
Production
Average Number of Elements
Mentioned
Adults mention 3-4 elements per
response.
• Similar elements regardless of
language
3.5
3
2.5
2
1.5
1
0.5
0
E-TD4
M-TD4
E-Adults M-Adults
Children mention between 1 & 2
elements.
• Language specific differences
are observed.
• English speaking kids are more
path oriented
• Mandarin speaking kids are
more goal oriented
Group: F(3, 62) = 53.188, p < .0001,
Adults > TD4, E-TD4 = M-TD4, E-Adults = M-Adults
M-TD4
E-TD4
M-Adults
E-Adults
1
Proportion of Element mentioned
4
Encoded Elments
0.8
0.6
0.4
0.2
0
Manner
Path
Goal
Source
Group: F(3, 62)= 53.179, p<.0001,
Adults > TD4
Encoded Elements: F(3, 62), p<.0001,
M > P > S, G > S
Group * Encoded Element: F(9, 62)= 6.891, p<.0001
Path: E-TD4 > M-TD4
Goal: M-TD4> E-TD4
Conclusion
• English and Mandarin adults encoded motion events similarly in both tasks and included
more elements than children.
• English and Mandarin speaking kids perceive these events similarly
• Mandarin kids encoded path less than English kids
• Our events do not easily lend themselves to serial verb constructions in Mandarin
• Reduced reliance on typical constructions may have affected results
Within Language Comparison:
Children with SLI & typically developing 4 year olds
Categorization Task & Attentional Biases
Do children with SLI notice the same things within the scene as TD children?
Children with SLI tend to focus on change of state rather than motion events
Kelly & Rice, 1994
Production Task
Given that children with SLI have a shorter MLU, what will they omit?
Omit Manner:
Heavy reliance on GAP verbs (e.g., go, do)
Rice & Bode, 1993 but see also Thordardottir & Ellis Weismer, 2001
Omit Source, Goal, Path:
Tend to omit optional arguments
Ingham, Fletcher, Schelletter, & Sinka, 1998; Johnston & Kamhi, 1984;
King, 1996, Thordardottir & Ellis Weismer, 2002
Participants
N
Age
CELF
ELI
SPELT
KBIT
TD4
SLI
19
4;8
(4;0 – 4;10)
6
108.89
(96 - 130)
111.26
(95-122)
104.68
(92 – 130)
SLI1
SLI2
SLI3
SLI4
SLI5
SLI6
6;5
5;3
5;8
5;9
5;10
8;0
8;4
78.67
89
83
95
63
75
67
80.33
78
76
73
76
97
82
92.67
110
80
89
83
97
97
Results
Categorization
Average Times Children chose the Mismatching Minimal Pair
12
SLI
TD4
10
8
6
Chi-squared test for different
independent groups
• All children fail to notice Source
changes or consider them irrelevant
SLI vs TD – All Elements
Χ2 (3) = 9.26, p=.02
• Differences in attention to Goal is
primary source of group differences
4
SLI vs TD Without Goal:
Χ2 (2) = 3.9, p=.16
2
• SLI children have a tendency to ignore
other changes if Goal changes
• TD children notice Manner & Path
changes more often
0
Mannar Bias
Path Bias
Goal Bias
Source Bias
Production
Avg. Number Items
Mentioned
4
3.5
SLI & TD4 talk both tend to mention between
1 & 2 elements
SLI group show a strong Goal Bias
•
•
•
•
3
2.5
2
The turtle went to the present
The giraffe got to the corn
The chicken wented to the swingset
The ant goed to the truck
Proportion of Responses that include Element
SLI
1
TD
0.8
0.6
0.4
TD4 group show a strong Manner Bias
1.5
1
0.5
0
SLI
TD
t(24)=.44, p=.66
•
•
•
•
A hippo was driving a car
A pig was flying
A fox riding in a jet over a rainbow
The deer jumped over the fence
0.2
0
Manner
Manner
Path
Path
t(24)=3.41, p=.002 Goal
t(24)=1.14, p=.26 Source
Goal
Source
t(24)=1.92, p=.06
t(24)=1.71 p=.10
Conclusion
• Children with SLI differentially attend to Goal instead of Manner and (to a lesser extent) Source
• Children with SLI rely heavily on GAP verbs, making it more likely that they mention goal
• Cannot disambiguate attentional bias from verb use given categorization results
• Groups are matched well in terms of amount of information included in the production task –
differences are not due to production limitations.
• Age-matched groups and more SLI children are required to validate the hypothesis that children
with SLI attend to different elements/talk about different aspects of the event
Conclusions
Cognition and Language in Motion Events
• Production & Categorization – Crosslinguistically
• Cross-linguistic production related differences were not reflected in the
categorization task.
- Easier to see production differences in children - below ceiling performance
- Need an intermediate age group to further explore language/ cognition
interaction
- Developmental changes in attention away from source
• Adults attend less to source and all groups encode source less in
production
- Cross-linguistic differences in the production task but not the categorization
task suggest that the changing attention to source is based on experience in
the world, not language alone.
• Production & Categorization -- Language Impairment
• Children with SLI attend to and talk about different things compared to their TD
peers.
- Following Kelly & Rice (1993, 1994), children with SLI may have different
attention biases, which change their verb learning profiles.
- Based on cross-linguistic results from the categorization task, we assume
no/limited covert linguistic encoding for the children.
• Both Children with SLI and Mandarin TD 4 year olds showed some evidence
of reliance on frequent constructions
• Follow up studies using more common actions and/or novel verbs required to
confirm this effect
Works Cited
Berman, R., & Slobin, D. I. (Eds.). (1994). Relating events in narrative: A crosslinguistic developmental study. Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Bowerman, M., & Choi, S. (2003). Space under construction: Language-specific spatial categorization in first language acquisition. In D. Gentner & S. Goldin-Meadow
(Eds.), Language in mind: Advances in the study of language and thought (pp. 387-427). Cambridge,
MA: MIT Press.
Choi, S., & Bowerman, M. (1991). Learning to express motion events in English and Korean: The influence of language-specific lexicalization patterns. Cognition, 41,
83-122..
Clark, E. V. (2004). How language acquisition builds on cognitive development. Trends in Cognitive Sciences, 8, 472-478.
Gennari, S., Sloman, S., Malt, B., & Fitch, W. (2002). Motion events in language and cognition. Cognition, 83, 49-79.
Ingham, R., Fletcher, P., Schelleter, C., & Sinka, I. (1998). Resultative VPs and specific language impairment. Language Acquisition, 7, 87-111.
Johnston, J. R., & Kamhi, A. G. (1984). Syntactic and semantic aspects of the utterances of language-impaired children: The same can be less. Merrill-Palmer
Quarterly: Journal of Developmental Psychology, 30(1), 65-85.
Papafragou, A., Massey, C., & Gleitman, L. (2002). Shake, rattle, ‘n’ roll: The representation of motion in language and cognition. Cognition, 84, 189-219.
Kelly, D. J. & Rice, M. L. (1993). Preferences for verb interpretation in children with specific language impairment. Journal of Speech and Hearing Research, 37, 182192.
King, G. (1996). Verb complementation in language impaired school age children. In M. Aldridge (Ed.), Child language (pp. 84–91). Clevedon, England: Multilingual
Matters.
Lin, S. (2006). Encoding motion events: A study of Mandarin-speaking children. Unpublished MA thesis. National Taiwan University, Taipei.
Papafragou, A., & Selimis, S. (2010). Event categorisation and language: A cross-linguistic study of motion. Language and Cognitive Processes, 25(2), 224-260.
Rakison, D. H. & Poulin-Dubois, D. (2002). You go this way and I’ll go that way: developmental changes in infant’s detection of correlation among static and dynamic
features in motion events. Child Development, 73, 682-699.
Rice, M. L., & Bode, J. V. (1993). GAPS in the verb lexicons of children with specific language impairment. First Language, 13, 113-131.
Slobin, D. I. (2004). The many ways to search for a frog: Linguistic typology and the expression of motion events. In S. Stro¨mqvist & L. Verhoeven (Eds.), Relating
events in narrative: Typological and contextual perspectives (pp. 219-257). Mahwah, NJ: Lawrence Erlbaum Associates Inc.
Talmy, L. (1975). Semantics and syntax of motion. In J. P. Kimball (Ed.), Syntax and semantics (Vol. IV, pp. 181-238). New York: Academic Press.
Talmy, L. (1985). Lexicalization patterns: Semantic structure in lexical forms. In T. Shopen (Ed.), Language typology and syntactic description. Vol. III: Grammatical
categories and the lexicon (pp. 57-149). New York: Cambridge University Press.
Talmy, L., (1991). Path to realization: A typology of event conflation. Proceedings of the Berkeley Linguistics Society, 17, 480-519.
Thordardottir, E. T., & Ellis Weismer, S. (2001). High-frequency verbs and verb diversity in the spontaneous speech of school-age children with specific language
impairment. International Journal of Language Communication Disorders, 36, 221-244.
Thordardottir, E. T., & Ellis Weismer, S. (2002). Verb argument structure weakness in specific language impairment in relation to age and utterance length. Clinical
Linguistics and Phonetics, 16, 233-250.
Acknowledgements
We would thank Karla McGregor, Bob McMurray, Word learning Lab and MACLab at the University of
Iowa for help and comments on the experimental design and stimuli. We also thank Hintat Cheung,
Lindsey Hansen, Allison Haskill, Elizabeth Lipton, Grantwood AEA, SungMei Prechool for all the help with
this project, and the members of Grammar Acquisition Lab at University of Iowa for data collection. This
project is funded by a Pre-doctoral Scholarship from Ministry of Education, Taiwan awarded to Shanju Lin,
and a University of Iowa Internal Funding Initiative awarded to Amanda J. Owen Van Horne.
Contacts: [email protected], [email protected]