Loading...
Please wait, while we are loading the content...
Similar Documents
Running Head: EARLY COMPREHENSION OF ACTION VERBS 1 Evidence for Early Comprehension of Action Verbs
| Content Provider | Semantic Scholar |
|---|---|
| Author | Nomikou, Iris |
| Copyright Year | 2018 |
| Abstract | Applying an eye-tracking technique, we tested early verb understanding in 48 infants aged 9 and 10 months. Infants saw two objects presented side by side and heard a verb that referred to a common action with one of these objects (e.g., eating relating to a banana). The verbs were spoken by the parent in an interrogative manner in order to elicit a looking behavior in the infant. Results showed that 9-month-old infants did not show recognition of our test words. However, 10-month-old infants were able to understand a number of the tested verbs. In the discussion, we relate our findings to the nature of early verb representations. Running Head: EARLY COMPREHENSION OF ACTION VERBS 3 Evidence for Early Comprehension of Action Verbs Studies on early language acquisition reveal accumulating evidence suggesting that 6month-old infants already understand language way before they speak their first words. Notably, these studies focus on the acquisition of nouns. One of the first studies used a preferential looking procedure and found that infants as young as 6 months of age understood some nouns such as mommy or daddy that were familiar from their everyday experience (Tincoff & Jusczyk, 1999). Follow-up studies revealed that at this early age, infants are also able to understand words referring to body parts such as hand and feet (Tincoff & Jusczyk, 2012) that are commonly used in early games between caregivers and their children. Bergelson and Swingley (2012) studied the comprehension of nouns in two types of categories: eight body parts (such as nose, mouth) and eight items related to food (such as spoon, banana). Using eye-tracking technology, they found that infants at the age of 6 months were not just capable of recognizing the correct item (e.g., banana) when presented paired on a screen with, for example, mouth. They could also select it from a more complex scene involving related objects on a table. Evidently, young infants are capable of referential word understanding before they can produce words. This understanding is not just interesting with respect to building up a lexicon. It may also foster the segmentation of the speech stream: Bortfeld, Morgan, Golinkoff, and Rathbun (2005) found that infants recognized novel words following familiar – but not unfamiliar – names. This effect might depend on infants’ knowledge of the meaning of the familiar word. Up to now, however, most studies on word comprehension have focused on infants’ early understanding of nouns. This is probably because researchers report that nouns are acquired earlier than verbs (e.g., Gentner, 1982; Gentner & Boroditsky, 2001). Indeed, research has shown that very few verbs appear in the first 10 (Tardif et al., 2008) or 50 words comprehended in either English (Fenson et al., 1994) or German (Kauschke & Hofmeister, Running Head: EARLY COMPREHENSION OF ACTION VERBS 4 2002). In other languages, it has been shown that verbs might occur earlier when that language provides a structure for their rich occurrence in input (see Au, Dapretto, & Song, 1994, and Choi & Gopnik, 1995, for Korean; Tardif, Gelman, & Xu, 1999, for Mandarin). Discussions on the development of semantics have considered whether early verbs differentiate between self-actions and observed actions. Investigating English-learning children, Huttenlocher, Smiley, and Charney (1983) conducted two studies: In a language comprehension study, 20to 42-month-olds were presented with 10 sets of two contrasting videos and asked to select one on which, for example, a person is running (in contrast to walking). The authors found support for their hypothesis that verbs relating to movements were the easiest for the children to identify, whereas verbs relating to change of involved entities (such as clean or take) were more difficult. To assess early spontaneous production, the authors analyzed recordings of 23to 28-month-olds in a variety of contexts (playground, store, and home). The data supported their proposed model of the development of action categories by showing that infants’ verb meanings for self-actions are acquired prior to meanings for observed actions; and that young children seem to analyze the latter in terms of movement. However, their model does not link up with findings presented in a more recent review by Sootsman Buresh, Woodward, and Brune (2006). This suggests that already: by the end of the first year of life, infants have the conceptual material relevant for a range of verb meanings, including not only verbs that encode the observable trajectories of moving objects (e.g., fall) but also verbs that encode the outcomes of causal sequences (e.g., open), intentional actions (e.g., get), transactions (e.g., give), and psychological states (e.g., see). (p. 214) Summarizing the literature on infants’ understanding of actions as well as early verb learning, these authors concluded that it is not just the physical motions through space that infants perceive when seeing other persons acting. Instead, from early on, infants understand human Running Head: EARLY COMPREHENSION OF ACTION VERBS 5 actions as relational with respect to the performer(s), objects, and goals. However, the authors also identified a gap indicating that it takes children one year to move from action representation to its productive expression. Semantic development during this phase is not well understood. The present study systematically investigated the age of acquisition of action verbs. We followed the approach presented by Mandler (2012), suggesting that early concepts capture the roles of objects—that is, what the objects do and what is done to them. Situated actions seem to be crucial for getting concepts about objects off the ground. This implies that actions themselves must be conceptualized early in development, and that these concepts form a solid basis for the acquisition of verbs. In addition, Heft (1989) postulates the relevance of bodily states that underlie the meaningful representations of entities. In contrast to nouns – which some approaches would argue involve one-to-one form– meaning mappings – verbs are relational, and their meaning involves the perception of a situation involving agents, actions, objects, and the relationships between them. Even though children seem to use nouns before verbs, it is possible that action information is crucial; and that in their initial use, nouns “cumulate” situational elements (Werner & Kaplan, 1963). For example, infants might say ball but relate it to the rolling ball (Nelson, 1974). Along similar lines, Samuelson, Smith, Perry, and Spencer (2011) found that children bind situational elements such as the location to an object’s name. Our study investigated 9and 10-month-old infants using a technique similar to that of Bergelson and Swingley (2012), except that we used verbs in place of nouns. In fact, in a study following up their work on understanding nouns, Bergelson and Swingley (2013) published an article on early understanding of abstract words in 6to 16-month-old infants: Among the pairs of words tested, three pairs consisted of action verbs (eat vs. hug, kiss vs. dance, and smile vs. drink); the others were mixed pairs (all gone vs. hi, more vs. splash, uhoh vs. bye, and wet vs. sleeping). These examples show that some of their tested words Running Head: EARLY COMPREHENSION OF ACTION VERBS 6 were more abstract than others, and the construction of the pairs did not focus on the syntactic role of the words. The results of this study revealed little understanding of any of the pairs by the 6to 9-month-old group, and the verb pair that infants in the 10to 13-month age range performed best on was kiss versus dance. The authors concluded that is not until the age of 14 to 16 months that evidence for understanding eat versus hug appears, although all seven pairs had positive scores. Thus, the data suggested tentatively that action verbs might be learned earlier than a variety of other word types such as all gone and hi. However, because the aim of the study was not to test action verbs alone, more work is needed to provide a clearer picture of the initial understanding of verbs. Instead of using dynamic pictures as in Bergelson and Swingley (2013), we used static object pairs and instructed the parent to ask for the relevant verb. Our study extends the Bergelson and Swingley (2013) design by using a setup including only verbs contrasted across contexts in a systematic way. Moreover, we used a much narrower age range. Method Participants We recruited 53 infants aged either 9 or 10 months. Five infants had to be excluded due to problems with calibration (2), hardware failure (1), and fussiness (2). Hence, the final sample contained 27 (16 female) 9-month-old (M = 9;7 [months; days], SD = 13 days) and 21 (12 female) 10-month-old infants (M = 10;11; SD = 11 days). Infants were recruited from the Bielefeld area in Germany. Recruitment took place mostly through personal contact with midwives. Through these midwives, it was possible to access mother–infant courses and address the mothers personally, inform them about the study, gather their contact details, and also leave flyers for other interested families to contact us themselves. All participants were full-term healthy infants with no history of any hearing impairment and they heard >80% German at home. In addition to participants’ demographics, we obtained some information on infants’ communicative as well as (loco)motor development via a parental survey that is Running Head: EARLY COMPREHENSION OF ACTION VERBS 7 currently under development (see Fischer, Nomikou, Grimminger, & Rohlfing, 2018). For this study, we considered only the information related to the child’s motor development. This is easy for parents to provide and therefore reliable. We simply asked whether or not the infants were already walking (regardless of whether this was in a supported or unsupported way). Materials Items. Motivated by |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://researchportal.port.ac.uk/portal/files/11296305/Nomikouetal_accepted.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
