Vienna, September 8, 2021 – Language acquisition is one of the most investigated topics of cognitive development, not surprisingly, considering that we still haven’t managed to fully understand how babies can acquire their language in such a fast and efficient way that adults would never be able to do so. One of the many questions, for example, often in the focus of research is how can babies make inferences regarding the meanings of novel words they hear in their environment. This is not an easy problem, because even in situations when they can rely on some further ‘helping cues’ – for example, when we point at a certain object while we say the new word – the number of potential meanings still remain countless. Who knows whether the novel word refers to a certain property of the object, or to some actions in relation to the object, or perhaps to our attitude regarding the object? Researcher Hanna Marno from CEU’s Department of Cognitive Science looked at this in a new study published last week by the American Psychological Association: Is 2-year-old children’s referential disambiguation of a novel word primarily aimed at identifying the word meaning or at understanding their communicative partner? - PsycNET (apa.org)
Researchers working on language acquisition processes have been trying to find the answer to this question. Due to studies conducted in the past decades, the research managed to shed light on some aspects of the mystery of word learning. For example, it turned out that in case the novel word is uttered in the presence of both a novel and a familiar object, the babies will ‘automatically’ associate the novel word with the novel object, that is, they don’t assume that the word would be a ‘second name’ or a synonym for the familiar object. Some researchers explain this phenomenon with the so-called ‘Novelty hypothesis’, according to which babies’ attentional processes are tuned in such a way that they tend to associate novel stimuli together automatically – in this case for example, the novel word with the novel object.
But would it only be a matter of low-level attentional processes how babies acquire novel words from their environment? Isn’t the main goal of language acquisition to be able to efficiently communicate with others, in other words, to understand what the other person is trying to convey so that we could answer in an adequate way? If so, then why babies wouldn’t have similar goals?
The study aims to find an answer to this question. In the experiment, a person first introduced a novel object to the baby and played with the object together with the baby for a few minutes before leaving the room, with the object remaining with the baby. Then another person entered who brought another three novel objects and placed them in front of the baby without interacting with her, and then she also left the room. At this point, the researcher let the baby play with all four objects for a few minutes on his own. Finally, the first person (with whom the baby was playing together at the beginning of the experiment) entered the room again and asked the baby to ‘put the room in order’ and to help her place all the objects on a tray. Once the baby put all the objects on the tray, she first pretended that she was about to leave the room again, but then suddenly stepped back to the baby and said: ‘Wait! Let’s leave the ‘Blicket’ here so that the other kids could play with it too! Could you help and put the ‘Blicket’ back, please?’ At this point, the critical question was: Which object would the baby choose?
According to the ‘Novelty hypothesis,’ the babies’ object choices should depend solely on the novelty on the objects, therefore in our context the baby should randomly pick one of the four objects (maybe with a bigger chance one out of the three objects that were brought in later, since they were relatively ‘more novel’ to the baby, compared to the first object, with which they had the common play). However, if the babies were trying to figure out the meaning of the word by answering to the question ‘Which object could she mean when she asked me to put back the ‘Blicket’?’, then most probably they would infer that the person must have referred to the first object, because that’s the only object they had a common experience together with.
The research result supported this latter assumption, as the majority of the babies indeed picked the first object in this context. Further control studies also gave evidence that the babies did not pick the first object only because they would have associated the object with the experimenter: when instead of asking for the ‘Blicket’ they were asked to pick ‘one toy’, they chose randomly out of the four objects. Similarly, they also tended to pick randomly out of the four object when instead of the first experimenter, a novel person entered the room and asked them to put back the ‘Blicket’, therefore we can also exclude the possibility that the babies object choices would have been a result of their subjective preferences related to the first object.
Based on these results we can conclude that when babies are trying to figure out the meanings of novel words, they do not rely only on low-level attentional mechanisms, but they are also highly motivated to understand the intentions of their communicative partner. Considering this, if we want to understand all the important factors that can contribute to babies’ efficient language acquisition, we should specifically focus on those pragmatic-social cues that we provide to them while we are using novel words in their presence.