At UTN, Prof. Dr. Charlotte Grosse Wiesmann investigates how babies learn and how their memory develops. In a recent study (“The self-reference memory bias is preceded by an other-reference bias in infancy”, Nature Communications, 2025), her team found that already within the first year of life, a baby’s social environment crucially shapes what they remember – they initially focus on what is important to others. In this interview, the professor explains how she came to study this topic, how research with babies is even possible, and why her findings offer insights not only for developmental psychology but also for artificial intelligence.
Professor Grosse Wiesmann, what personally led you to study learning and memory in babies?
I originally focused on social cognition – how children learn to take another person’s perspective. I noticed that even very young children are strongly influenced by others in both their behavior and memory. Although they don’t yet fully grasp that others may see the world differently, other people’s reactions already shape what they remember. That phenomenon led me to the topic of memory.
What exactly did you want to find out in your study about babies’ memory?
We wanted to find out when children begin to distinguish between what matters to others and what is meaningful to themselves. Surprisingly, we discovered that during the first year of life, babies mainly remember things that are important to others. Only around 18 months – when a first sense of self begins to form – does the focus shift toward personally relevant experiences.
What does this reveal about the role of relationships and social context in early childhood?
Our findings highlight how deeply other people shape a baby’s development. Babies initially cannot structure what’s important in their complex environment, they need caregivers to do that. These interactions help them assign meaning amid the flood of impressions and to prioritize what they learn and remember.
How can you actually conduct studies with babies?
We designed novel toys that the babies had never seen before. Each object was assigned either to themselves or to another person, namely a puppet named Sven. Later, we tested whether the babies remembered the objects. Using eye-tracking, we measured where they looked and for how long, which told us which objects they recognized.
What was the most surprising result?
That babies in their first year mainly remember what is important to others. They didn’t seem to remember objects that were only meant for themselves. Later on, as their self-concept develops, children increasingly remember self-relevant experiences.
You emphasize that studying how children learn can also inspire artificial intelligence. What fascinates you about that connection?
Babies learn extremely fast, but never alone. Other people help them identify objects in their environment and assign meaning. This social guidance directs their attention and helps them distinguish what’s important from what isn’t. That’s exactly what artificial intelligence currently lacks: it processes huge amounts of data, but without any social scaffolding or guidance. By understanding how babies use these “attentional filters,” we might teach machines to learn more efficiently and more like humans.
The puppet Sven helps researchers better understand babies’ behavior.
You joined UTN last year. What research topics will you pursue here?
I’m particularly interested in how babies form complex categories so rapidly and how their social cognition develops. I also work across disciplines with colleagues in computer science and philosophy, for example, to explore how early learning can inspire new approaches in AI and robotics.
What can students interested in this field expect from you?
In the winter semester 2025/2026, I’ll launch an introductory course on empirical research, where students will design their own small experiments. Later, there will be advanced seminars on social cognition and learning. Students can also do internships or write their theses in my research group, and there will be exciting opportunities for PhD and postdoctoral projects.
The research of Prof. Charlotte Grosse Wiesmann shows: Even in the first year of life, learning is a profoundly social process. Babies do not remember things at random but their attention is guided by what others find important. Only later does their own self take center stage. This delicate interplay between relationships and memory underscores how essential social bonds are for development and offers inspiration for artificial intelligence. Perhaps one day, machines will learn more like children do: not through massive datasets alone, but through intelligent social guidance.
At UTN, Prof. Dr. Charlotte Grosse Wiesmann investigates how babies learn and how their memory develops. In a recent study (“The self-reference memory bias is preceded by an other-reference bias in infancy”, Nature Communications, 2025), her team found that already within the first year of life, a baby’s social environment crucially shapes what they remember – they initially focus on what is important to others. In this interview, the professor explains how she came to study this topic, how research with babies is even possible, and why her findings offer insights not only for developmental psychology but also for artificial intelligence.
Professor Grosse Wiesmann, what personally led you to study learning and memory in babies?
I originally focused on social cognition – how children learn to take another person’s perspective. I noticed that even very young children are strongly influenced by others in both their behavior and memory. Although they don’t yet fully grasp that others may see the world differently, other people’s reactions already shape what they remember. That phenomenon led me to the topic of memory.
What exactly did you want to find out in your study about babies’ memory?
We wanted to find out when children begin to distinguish between what matters to others and what is meaningful to themselves. Surprisingly, we discovered that during the first year of life, babies mainly remember things that are important to others. Only around 18 months – when a first sense of self begins to form – does the focus shift toward personally relevant experiences.
What does this reveal about the role of relationships and social context in early childhood?
Our findings highlight how deeply other people shape a baby’s development. Babies initially cannot structure what’s important in their complex environment, they need caregivers to do that. These interactions help them assign meaning amid the flood of impressions and to prioritize what they learn and remember.
How can you actually conduct studies with babies?
We designed novel toys that the babies had never seen before. Each object was assigned either to themselves or to another person, namely a puppet named Sven. Later, we tested whether the babies remembered the objects. Using eye-tracking, we measured where they looked and for how long, which told us which objects they recognized.
What was the most surprising result?
That babies in their first year mainly remember what is important to others. They didn’t seem to remember objects that were only meant for themselves. Later on, as their self-concept develops, children increasingly remember self-relevant experiences.
You emphasize that studying how children learn can also inspire artificial intelligence. What fascinates you about that connection?
Babies learn extremely fast, but never alone. Other people help them identify objects in their environment and assign meaning. This social guidance directs their attention and helps them distinguish what’s important from what isn’t. That’s exactly what artificial intelligence currently lacks: it processes huge amounts of data, but without any social scaffolding or guidance. By understanding how babies use these “attentional filters,” we might teach machines to learn more efficiently and more like humans.
You joined UTN last year. What research topics will you pursue here?
I’m particularly interested in how babies form complex categories so rapidly and how their social cognition develops. I also work across disciplines with colleagues in computer science and philosophy, for example, to explore how early learning can inspire new approaches in AI and robotics.
What can students interested in this field expect from you?
In the winter semester 2025/2026, I’ll launch an introductory course on empirical research, where students will design their own small experiments. Later, there will be advanced seminars on social cognition and learning. Students can also do internships or write their theses in my research group, and there will be exciting opportunities for PhD and postdoctoral projects.
The research of Prof. Charlotte Grosse Wiesmann shows: Even in the first year of life, learning is a profoundly social process. Babies do not remember things at random but their attention is guided by what others find important. Only later does their own self take center stage. This delicate interplay between relationships and memory underscores how essential social bonds are for development and offers inspiration for artificial intelligence. Perhaps one day, machines will learn more like children do: not through massive datasets alone, but through intelligent social guidance.
Read the study online
Learn more about the Cognitive Neuroscience research group