Authors: Alex de Carvalho, Victor Gomes and John Trueswell
Contact Email(s): alex.de-carvalho@u-paris.fr, vgomes@sas.upenn.edu, trueswel@psych.upenn.edu
7 thoughts on “English-learning preschoolers use negative sentences to constrain novel word meanings”
Leave a Comment or Question Below
You must be logged in to post a comment.
Great talk, thanks Alex! I was wondering, could you use “this is wug a blicket” along with negative non-linguistic cues to tease apart the role of linguistic and non-linguistic cues?
That is an interesting suggestion, Julie! Thanks! I will think more about it.
My guess is that if we simply replace “not” by “wug” in the exactly some context of our teaching phase, children will fall into one of the scenarios below:
1) They will be confused or behave like children who does not know how to compute the meaning of sentences with negation yet
2) Maybe they will have to learn that “wug” is a kind of novel function word that has something to do with negation or contrasting information (e.g., different) and you saw that they can exploit the semantic context to learn novel function words in Anne’s talk, right?.
3) There is even a worst (but unlikely) case scenario that is they might parse “wugablicket” as an entire new word. But of course, we can avoid that by imposing the same kind of prosody we had on the real negative functor “not” on the novel functor “wug”.
For the negation group the test trial 2 was presenting a conflict. Mutual exclusivity would lead participants to reject both objects as blickets: the not-blickets are no blickets but the novel objects are clearly different from the previously presented blickets (to the same degree with regard to perceptual differences as the not blickts). So is it fair to compare children and adults on such a difficult condition? While adults might select the novel object might employ some additional reasoning, children might take mutual exclusivity more “seriously” than the negation and select the ‘not-blicket’. So the difference in performance between adults and children would be explained by this fact rather than by children having more difficulty with negation. (Just to be clear: This does not take away from the main finding that children did use negation refine the meaning of the newly learned word.)
Hi Tom, Thanks a lot for your thoughts about our results. Yes I agree that there are other ways to explain why children behaved differently from adults in our “test 2 trials”. We indeed don’t think that they have necessarily more difficulty with negation per se, but certainly that the additional inferences that they had to do in “test 2 trials” were more difficult for them as you say in your comment.
Maybe adults are better in understanding the pragmatic context of the experiment than children. Let’s say, adults in the negative condition inferred that the unrelated object was a “blicket” simply because at test, the verbal prompt “Do you see the blicket?” implied that there was a blicket on the screen. So, adults might in this situation have used a “mutual exclusivity” strategy, and reasoned that since the “pinkish monster” was not a blicket, then only the unrelated character could then be a blicket. Children might have had difficulty in doing all these additional inferences. That is why I said in the discussion that maybe it’s not a difficulty in interpreting negative sentences per se (e.g., A vs not A), but rather a problem in using the restrictive information to make additional inferences in certain tasks (e.g., if not A, then B).
Thanks for the nice talk, Alex, very interesting results! Partially related to Roman’s question, are you thinking about running this with younger children in a preferential looking paradigm? I’m wondering if the evidence for age of comprehension could be pushed down to somewhere in the second year of life.
I guess Roman’s question would still stand, since presumably there is some age at which kids don’t understand negation. Perhaps a filtering story would then be useful, like the one Laurel Perkins gives for WH questions and the acquisition of argument structure.
Hi Daniel,
Yes we might certainly try to do this experiment with younger children when all this covid crisis is over.
In the meantime, you might want to have a look at a paper we recently published showing evidence that 18-month-olds are able to understand negative sentences in French and can potentially use this information to constrain their interpretation of word meanings.
The ref is the following:
de Carvalho, A., Crimon, C., Barrault, A., Trueswell, J., & Christophe, A. (2021). “Look! It is not a bamoule!” 18- and 24-month-olds can use negative sentences to constrain their interpretation of novel word meanings. Developmental Science.
I am happy to send you a copy of the paper if you email me. 😉
Nice, I just found it on your website. This is what I was expecting!