How do children learn to talk about the world around them? My past work has investigated how children acquire nouns (Hirsh-Pasek, Hennon, Golinkoff, Pence, Pulverman, Sootsman, Pruden, & Maguire, 2003; Hirsh-Pasek, K., Golinkoff, R.M., Hennon, E. A., & Maguire, 2004; Maguire, Hennon, Hirsh-Pasek, Golinkoff, Slutzky, & Sootsman, 2002; Maguire, Hennon, Slutzk, & Golinkoff, 2001) and particularly how they acquire verbs (Golinkoff, Chung, Hirsh-Pasek, Liu, Bertenthal, Brand, Maguire, & Hennon, 2002; Maguire, 2007; Maguire & Dove, 2007; Maguire, Hirsh-Pasek, & Golinkoff, 2007; Maguire, et al., 2004; Maguire, et al, 2003; Pruden, Hirsh-Pasek, Maguire, & Meyer, 2003). One of the most consistent findings in language development is that verbs are acquired later than nouns (Clark, 2003; Imai, Haryu & Hiroyuki, 2003). I have argued, based both on findings in language development (Gentner & Boroditsky, 2001; Imai, Haryu & Hiroyuki, 2003; Meyer, et al., 2003; Smiley & Huttenlocher, 1995) and on neurocognition (Bird, Howard, & Franklin, 2003; Black & Chiat, 2003), that this different might not be solely a problem between nouns and verbs but a difference in the ability to form a mental image of the item, sometimes called imageability or concretness (Maguire, 2007; Maguire & Dove, 2007; Maguire, Hirsh-Pasek, & Golinkoff, 2007). I’ve remained agnostic on the arguments between imageability and concreteness arguing that the data currently isn’t strong enough to differentiate the two (Maguire, Hirsh-Pasek, & Golinkoff, 2007). My current work now attempts to use neurocognition to help determine how differences in processing imageability and grammatical class change over time
My ongoing research at the Callier Center on adult language processing with ERPs has reported differences in processing for nouns compared to verbs and for high imageable words compared to low imageable words (Druks & Masterson, 2003; Kounios & Holocomb, 1994; Swaab, et al., 2002). A few studies have looked at the interactions between these. For example Kellenbach, et al (2001) reported that adults showed differential brainwaves between nouns at verbs at two points, between types between 250-350 ms and again in a later portion of the wave between 350-450 ms. Further they showed differences between high and low imageable words between 250 and 750 msec. The authors relate these to the P2 and N400 peaks, yet no interaction between grammatical class and imageability processing.
To date no one has studied this in children, though many of our theories of language development related to either the grammatical class, the imageability (or similar cognitive construct such as concreteness) or some mix of the two. We are currently using ERPs to study how processing of imageablity and grammatical class differ in adults and specifically how they develop in children.
A second study investigates children’s ability to necessary grammatical extension to novel verb types. Research with adults consistently shows two distinct ERP components that differ when processing grammatically correct or incorrect sentences (the LAN (or ELAN) and the p600 (see Hahne & Friederici, 1999 or Friederici, 2002 for review). Further there is evidence that adults show these same differences when listening to jabberwalky
sentences, which are thought to have grammatical properties but little or no semantic information. For example, The snipe grambled herpishly
vs. Snipe the grambled herpishly
(Hahne & Jescheniak, 2001). It has been argued that differences in the adult brainwaves in relation to these errors indicate that grammatical processing can be independent of semantic processing. Our interests are only how children detect errors in novel verbs and how this changes over time. As a result we are testing the development of processing these grammatical errors with known verbs, verbs whose meanings are just learned in a different verb form (teaching the meaning of to lorp
and testing on errors in lorping
), and verbs whose meanings are unknown (testing on errors in lorping
though that word has never been learned). Our goal is to uncover when neurological processes are in place for children to detect grammatical violations to novel verbs and how that changes over time. Primarily we are tracking the development of the ELAN compared to the p600 to help uncover the levels of grammatical processing across development.
In addition to work in neuroscience, I have continued to expand my work in language acquisition using behavioral measures. To this end investigations into how different languages, and in some cases multiple languages, influence cognition and development. Here we have two projects underway.
The first project is in collaboration with Mutsumi Imai in Japan. We are investigating differences in children interpretations of a novel verb based on their native language. Languages differ in their verb system in terms of the information often encoded. For example, in English speakers often encode the manner of the action, or the way in which the actor moves (run, jog, walk, swagger, hop, skip, jump). In many other languages including Japanese and Spanish, the path of the action, or agent trajectory, is often encoded in the head verb (ascendere). Our research investigates how this changes children’s perception of novel verb labels.
A further project, lead by my graduate student, Sandra Vengas, investigates the interaction of theory of mind and language acquisition in bilingual children.