What we have found out so far
Listening and Speaking: How Children's Brains Process Language
Principal Investigator: Tiziana Srdoč
Project Duration: 2024-2026
Language is a complex skill, and different areas of the brain must work together so that we can both understand and speak words. While we already know quite a bit about which regions of the brain are active in adults when they hear or speak words, it is still unclear how the child's brain performs these tasks.
Some studies have shown that school-age children often process language in both hemispheres of the brain, unlike adults, in whom the left hemisphere plays a much greater role. In our FIRST study, we want to find out which areas of the brain support speech and word comprehension in three-year-old children.
To investigate this question, we used a special method: functional near-infrared spectroscopy, which uses light of a specific wavelength to measure the oxygen concentration in different areas of the brain. This allows us to observe which regions are currently “working.” In the study, the children looked at various pictures of animals and objects and heard the names (e.g., “cat”) or named the pictures themselves. Sometimes we deliberately showed them distorted images, which they were asked to describe as “nonsense,” or distorted, incomprehensible sounds. This control condition helped us to distinguish which brain regions are active in simple perception and which are involved in object recognition and language.
Our findings confirm that children use both sides of their brain when naming pictures: the typical language regions of the left hemisphere, but also certain areas of the right hemisphere. When listening, the part of the right hemisphere responsible for understanding melody and speech rhythm was particularly active. This means that three-year-olds still use a broad, bilateral network when speaking and listening. Their brains are therefore not yet as specialized as those of adults.
These findings are consistent with the theory of “neural efficiency.” Younger children use larger and less specialized areas of the brain while their language systems are still developing. As they get older, processing becomes more efficient and focused. Studies like these help us to better understand language development and identify language delays early on.
Master Thesis “Lighting up the developing brain” on u:theses
How to interpret finger pointing
Principal Investigator: Tibor Tauzin
Project Duration: 2024 -
Pointing as a communicative gesture emerges early in development. Infants do not simply produce it to draw others' attention to something important; they can also understand when others point. However, pointing can indicate an important object that we want to show to others. At the same time, it can also provide information about the location or direction of a relevant place.
For instance, in a context where we point at an object and name it (“Look, it’s a ball”), pointing serves as a cue to highlight the object. In other cases, for example, when someone points toward an object inside a box, we can understand that they are not referring to the box itself but to the object inside it. Thus, pointing has two different meanings.
Historically, it was assumed that infants’ default interpretation of pointing is that it refers to an object rather than a direction. However, this question was never directly tested. Therefore, we designed the study Meaning of Pointing ("MoP) to test whether infants, indeed, interpret pointing as referring to an object.
To test this prediction, 14-month-old infants were presented with two objects. Subsequently, one of the objects was pointed at and then swapped with the other (non-indicated) object. We hypothesized that if infants understand pointing as referring to an object, they should choose the one that was pointed at, irrespective of the fact that it is now in a new location. However, if infants interpret pointing as referring to a direction or location where an important or interesting object can be found, they should choose the non-indicated object now located in the indicated direction.
We found evidence showing that infants interpret pointing as indicating a direction rather than an object. These results align with our previous findings, which show that other species exhibit a similar tendency. This suggests that how humans interpret pointing may have an evolutionary ancient origin and that the canonical interpretation of pointing (as referring to objects rather than places) may follow an independent developmental trajectory. Importantly, since it is rare for an object to be replaced by another just after pointing, infants likely rely on pointing to learn about objects. However, it may take time for them to understand that pointing can serve different purposes.
Do you recognize this song? "Kinderlieder" study with 7 month old babies
Principal Investigators: Anja-Xiaoxing Cui, Jutta L. Mueller, Ingrid Schachner, Ivonne Weyers
Duration: 2024 -
Babies hear songs all over the world. The extent and the lyrics of the songs may differ, but all nursery rhymes have one thing in common: linguistic and melodic elements are combined. As part of our Kinderlieder study, which was a cooperation between Anja Xiaoxing Cui (Institute of Musicology) and the Babelfisch Lab, we wanted to investigate which elements are particularly interesting for babies and whether they use these elements to recognize songs.
A total of 100 seven-month-old babies took part in our nursery rhyme study. At home, the families listened to a song composed especially for the study. In the Babelfisch laboratory, the babies where presented the same song again and also another, unknown song. Both songs were composed in such a way that the linguistic and melodic elements were of different lengths. Using the brain activity measured by EEG during the laboratory visit, we are studying whether these different lengths are also reflected in the brain activity. An initial, preliminary analysis has now shown us that the length of the bars, i.e. a melodic element, is more likely to be represented than the length of the notes, another melodic element. This result alone is very exciting, as it indicates that babies can already recognize bars without knowing what a bar is! We are very much looking forward to the results of further upcoming analyses and will report as soon as these have been completed.
Many Babies 3 study: Do 7-12 month old babies discover rules in syllable sequences?
Principal Investigator: Ivonne Weyers
Project Duration: 2023-2026
Even before babies say their first word, they have already learned quite a bit about their native language(s): they can identify individual, recurring words and even recognize structural relationships between words. As part of “Many Babies 3”, a large-scale international study was investigating exactly how babies discover certain rules in simple syllable sequences. More than 40 developmental laboratories worldwide, involving 650 babies, participated in this study. With such a large number of participants, the researchers hope to obtain particularly meaningful results.
In the Babelfisch laboratory, we tested 40 babies aged between 7 and 12 months by measuring their eye movements. In a kind of “learning phase”, the babies first heard fantasy words that followed a certain rule: babies in group A heard fantasy words in the sequence ABB (e.g. lisoso, metata), babies in group B heard fantasy words in the sequence ABA (e.g. lisoli, metame). The children then heard new examples that either corresponded to the previously heard rule or deviated from it. Eye-tracking was used to measure which of these examples the babies paid attention to for longer. Did they look longer if they thought what they heard was familiar/new? This would be an indication that they had noticed and learned the previously presented rule. Or did they show no preference and looked for about the same length of time?
Until now, it was assumed that babies learn abstract rules very quickly when they are played for them in the form of syllable sequences. The ManyBabies3 study has now shown that this is not necessarily so easy: No evidence was found of learning an abstract rule based on a pattern of repetition.
Link to the publication ManyBabies 3: A Multi‐Lab Study of Infant Algebraic Rule Learning
"MOUTH" study: Do 7-9 month old babies imagine speech sounds?
Principal Investigator: Sandra Regen
Duration of Project: 2023-2024
Within the first few months of life, babies already gain a lot of experience with the audiovisual perception of speech whenever they not only hear people around them speak, but also watch them speak. In addition, babies begin to babble themselves from around the age of 7 months and thus gain their first experience with the production of speech sounds - and with how they sound and exactly what movements they produce. One unanswered question that concerns the Babelfisch team is whether babies can not only perceive and produce speech sounds in detail, but whether they can also imagine them in their “mind's ear”.
In our “MOUTH” study, we used eye-tracking to examine the babies' attention during the following task: first, they were presented with a silent video showing the face of a person uttering a speech sound (an “a”) or a non-speech sound (vibrating lips). Then only one sound was heard, without the video. The decisive factor was that the audible sound matched the previously shown video in half of the trials and not in the other half. Would the babies look for different lengths of time depending on whether the sound matched the video or not? If this were the case, it would be an indication that they had already imagined the linguistic or non-speech sound in their “mind's ear” when they only saw the mouth movement of the sound.
However, the results of our study with 7-9 month old babies could not prove such differences. Therefore, we cannot yet draw any conclusions as to whether babies imagine speech sounds. However, our results show that babies look attentively at the screen for longer when listening to speech sounds than when listening to non-speech sounds. Speech therefore always seems to be more interesting than other sounds, regardless of whether its occurrence was announced by visual information or not.
We will continue to work on this exciting question in the future to find out whether and, above all, from when babies can imagine speech sounds!
Master thesis “Covert activation of speech sounds in preverbal infants?” on u:theses
