Three scientific studies shed new light on early learning processes

If you have a close relationship with a kid, your own or someone else's, you may have created personal ways to bond with them, taught them the names of objects, and helped them to read and, ultimately, to grow. When you spend time with a kid, you develop a feeling for the stories and songs they like, the animals they prefer, and the kind of environment that helps them fall asleep — long before the child has matured to the point of verbal communication.

Those feelings are often the result of many hours of conscious and subconscious observation, the kind of closeness parents develop that helps them to nurture and foster development and learning in their kids. Our innate ability to learn is a critical life essential because it directly impacts our quality of life. And because of this, scientists have dedicated massive amounts of time and research effort to deepen our understanding of infant learning processes — in the hope that we can identify and resolve difficulties early on and develop innovative and inclusive learning solutions.

Unfortunately, manual observation is subjective at best, expensive, and unsystematic. In addition, young babies and infants haven't developed the skills to articulate their thoughts and feelings accurately, making it challenging to research early learning processes systematically.

And that's where attention computing solutions come into play. This technology can accurately measure gestures — even microscopic eye movements and blinks — without invading our natural environment. In this post, we highlight three separate infant studies that have leveraged attention computing to shed new light on early childhood development.

1. Babies develop caring capabilities by staring at your face

You have probably noticed that babies have a tendency to stare — even at strangers. While sitting on a bus or waiting in line, you may have captured the imagination of a young baby who will stare at you for what can sometimes feel like an uncomfortable amount of time — certainly longer than what is deemed appropriate in many cultures.

What makes babies do this? Are they simply trying to process the image in front of them, or is a more profound activity taking place? Most parents will instinctively say that they can feel the intensity of their child's mind and the development that's taking place when their children stare. And recent research shows there is more to a baby's stare than figuring out whether the person they are looking at is happy or sad.

We know this because a group of Finnish researchers (Mikko J. Peltola, Santeri Yrttiaho, and Jukka M. Leppänen) tasked themselves with measuring babies' attention bias to faces to determine if there is a correlation between this bias and caring behaviors developed during infancy. The researchers followed a group of children through the first years of childhood, using attention computing to measure attention bias to faces in babies at about seven months.

Bubbles indicate fixations — the larger the bubble, the longer the fixation. This image is a simplified version of typical eye movement patterns (not scientifically accurate or based on actual data).

The researchers first uncovered that babies possess a distinct attention bias for faces — especially when met with fearful expressions. The study also showed that this bias varies from one kid to the next and declines as they develop. And they discovered a correlation between strong attention bias for faces with deeper helping responses at two years and fewer callous-unemotional traits at four.

So, the next time you find yourself face-to-face with a six-month-old baby, smile and think about how you are helping this kid to develop.

More about this research, funded by the Academy of Finland and the European Research Council, is available here.

2. Children learn new animal names easily if they like animals

Irrespective of mother tongue, it seems that the processes children across the globe use to learn and remember words are similar, with comparable vocabulary development. However, the terms a child learns — such as cat, dog, or pasta — vary drastically from one kid to the next. We have rationalized this variation by assuming that children are exposed to words at different times during development. But new research leads us in another direction — that the child's interest in the object category shapes their capacity for new words. So, for example, kids who like animals will find it easier to learn new animal names than objects that belong to a category that doesn't interest them, such as flowers or vehicles.

This new research was carried out on a group of 30-month-old monolingual German-speaking kids who were full-term with regular hearing and vision. To set a baseline, the researchers used attention computing to assess each child's level of interest in different types of objects — animals, clothes, drinks, and vehicles. By showing the kids word-object association cards, the attention computing solution could assess their interest based on pupil dilation measurements.

In a second test, the researchers exposed the kids to a different set of word-object cards, enabling them to assess how well each child could learn and recall new words. They found that kids exhibit robust learning if they are interested in the object category and that personal passion and excitement empower learning.

The complete paper by researchers Lena Ackermann (University of Goettingen), Robert Hepach (Leipzig University), and Nivedita Mani Leipzig University, is available here.

3. Kids learn faster when the task involves social interaction

It makes sense for most parents and caregivers that social interaction encourages infant learning. With the help of attention computing, recent research (funded by Leipzig University and the Max Planck Society for the Advancement of Science) has shown this to be the case. Interestingly, the study reinforces our belief that spending face time with children helps them develop. But, it also revealed that situations where people look at each other when communicating stimulate learning.

The attention computing solution used in the study revealed faster saccadic rhythm and more predictive gaze shifts — movements that are indicative of an intense level of learning — in scenarios that included face-to-face social interaction.

The complete paper by Maleen Thiele, Robert Hepach, Christine Michel, Gustaf Gredebäck, and Daniel B. M. Haun is available here.

Attention computing — is the convergence of sensor technologies that enable people and machines to communicate in much the same way as humans interact with each other. It is an advanced level of human-computer interaction, augmenting traditional device-based input methods like mice, keyboards, and touchscreens with voice, gestures, facial expressions, and movement. It relies on computer vision, machine learning, and natural language processing to translate human reactions into high-level attention states such as cognitive load, focus area, eye openness, and pupil dilation. An application can use these insights to decode what captures a person's attention (and what doesn't) as well as their intent. Attention computing aims to remove the barriers between people and machines, to make technology accessible to everyone, and promote the democratization of services.

Parents or technology — who knows best?

If we accept that people who spend a lot of time with a child gain a good sense of what the child needs to grow, then harnessing that knowledge will help us develop innovative and inclusive learning solutions that scale. The three studies outlined in this post show that attention computing can help developmental researchers to do just that. To understand kids — especially those yet to develop verbal skills — without intruding on their environment or affecting their behavior.

So, while parents probably know what's best for their own kids, technology enables researchers to capture that knowledge systematically, making it useful for everyone.

One company that has done just that is BrainLeap — a California-based tech startup on a mission to unlock the potential of more than one million children with attention challenges — every year. They have built a solution that leverages their own research about the correlation between eye movement and attention, creating an attention training game that has proven to raise attention skills with a positive impact on overall learning capabilities in young children.