Processing Events and Learning Relational Terms: Figures are More Prominent than Grounds

Transcription

1 Processing Events and Learning Relational Terms: Figures are More Prominent than Grounds Tilbe Göksun Kathy Hirsh-Pasek Sarah Roseberry Temple University Roberta Michnick Golinkoff University of Delaware XVIth International Conference on Infant Studies Vancouver, 2008

2 To process events and learn relational terms (e.g., verbs and prepositions), infants must: Parse the events into components Perceive and categorize these components Abstract them in the dynamic events Package and lexicalize the components according to their native language Gentner, 1982; Gentner & Boroditsky, 2001; Golinkoff et al.,2002; Hirsh-Pasek & Golinkoff,

3 Jackendoff, 1983; Langacker, 1987; Lakoff, 1987; Mandler, 1996; 2004 Talmy, 1985;

4 These semantic components are Perceptually accessible Mandler, 1996; 2004 Universally codified in languages Jackendoff, 1983; Langacker, 1987; Talmy, 1985 But also encoded differentially in languages e.g. Path-Manner English - Manner encoded in verb; path encoded in preposition (e.g., climb up). Spanish, Turkish, Greek - Path encoded in verb; manner encoded in adverb (optionally) (e.g., in Turkish: tirmanarak cikti going up climbingly ). Conceptual primitives Jackendoff, 1983; Lakoff, 1985; Mandler,

5 Path - Manner Discrimination of Path-Manner 7-month-olds discriminate path and manner in animated scenes Pulverman et al., 2003; month-olds discriminate path and manner Naturalistic scenes and human agents Casasola, Hohenstein, & Naigles,

6 Path - Manner Categorization of Path-Manner 11-month-olds categorize path over different manners (e.g., jump over, twist over, flap over) Pruden et al., 2003; months-olds categorize manner over different paths (e.g., jump over, jump under, jump behind) Pruden et al., 2003; month-olds categorize manner across changes in actors and paths in realistic stimuli Song et al.,

7 Path - Manner When infants start learning language: Pay attention to language-specific aspects 14- to 17-month-old English-reared infants having higher vocabularies are more sensitive to manner Spanish-reared infants at the same age who have lower vocabularies are more sensitive to manner Pulverman et al., 2003 English, Japanese, and Spanish speaking toddlers: extend the novel verb to label path by preschool age, children display language specific preference Maguire et al., in progress 7

8 So Infants parse the semantic components of actions Categorize the components Developmental progression: find certain constructs easier than others Be sensitive to their native language s encoding with exposure 8

9 Figure is the moving entity performing the action Ground is the stationary setting where the action occurs e.g., The girl climbs up the mountain 9

10 Figure - Ground Universally codified in languages Jackendoff, 1983; Langacker, 1987; Talmy, 1985 Language specific encoding Ground encoding English Go into refers to a path that the figure moves along and the ground object as some kind of enclosure Japanese: ground path verbs ---- ground is incorporated into the verb wataru go across A barrier between a starting point and an endpoint The ground should be a flat extended surface The man go across the road/bridge, but not mountain, tennis court or grass. Muehleisen & Imai, 1997; Tsujimura,

11 Present Study Do pre-linguistic infants discriminate figures and grounds in non-linguistic dynamic events? Is there primacy for figure or ground in processing these events? Are English-reared infants sensitive to the ground distinctions/comparisons encoded by Japanese verbs (e.g., wataru)? 11

12 Method Participants Study 1: Figure Discrimination 7-9 month-olds (N= 24, M= 7.89 mo) month-olds (N= 14, M= mo) Study 2: Ground Discrimination 7-9 month-olds (N= 23, M= 8.15 mo) month-olds (N= 14, M= mo) 12

13 Method Preferential Looking Paradigm Nonlinguistic dynamic events DV: Looking Time Example 13

14 Method 14

15 Study 1: Figure Discrimination Familiarization Test 15

16 Specific Predictions No preference to either side at salience Age differences Preference to novel figure at Test Condition differences Adult-Adult, Child-Child, and Adult-Child Figures on different grounds 16

17 Results 17

18 Specific Predictions No preference to either side at salience Age differences YES Preference to novel figure at Test NO: At 7-9 months YES: At months Condition differences NO, but adult-child conditions NO difference for conditions having different grounds 18

19 Study 2: Ground Discrimination Example 1: In Category Familiarization Test 19

20 Study 2: Ground Discrimination Example 2: Out Category Familiarization Test 20

21 Specific Predictions No preference to either side at salience Age differences Preference to novel ground at Test Condition differences Differences in ground categories 21

22 Results 22

23 Results 23

24 Results e.g., railroad - road e.g., railroad tennis court 24

25 Specific Predictions No preference to either side at salience Age differences YES Preference to novel ground at Test NO: At 7-9 and months YES: Only at months Condition differences Differences in ground categories YES: Differentiated grounds depending on ground-path verb categories (13-15 months) 25

26 Do infants differentiate perceptually salient grounds? Out category In category 26

27 Control experiment Black & White Screen 27

28 Together Infants discriminate figures and grounds in nonlinguistic dynamic events by 12 to 15 months of age, respectively Figures are distinguished earlier than grounds When English-reared infants start discriminating grounds in nonlinguistic dynamic events, they pay attention to differences not typically codified in their language 28

29 Discussion Discrimination of grounds Categorical distinctions in Japanese Infants start with paying attention to narrow differences Parallels to phonological development Discrimination of figures moving entity Infants pay attention to motion Findings are parallel to other semantic components in the literature Differential attention to some aspects of events over others (path before manner, figure before ground) 29

30 Future Studies Older child age groups and adults Parallel studies with Japanese-reared infants Context vs. ground Bornstein et al. (2007) Comparison of grounds with a different action walking alongside the road Categorization Do infants categorize grounds depending on their category relation? 30