One of the first things a newborn does is track faces. Within minutes of birth, infants will turn their heads toward simple arrangements of dark shapes that resemble eyes and a mouth. This early sensitivity is not learned. It is a built-in system that gives the infant a starting point for understanding the world.

Mark Johnson and his colleagues call this system CONSPEC. It guides the newborn’s attention to faces during the first weeks of life. In Mastery, I described this as the “primal sketch” of human vision. These early fixations begin a perceptual cascade that helps humans lean into a world filled with visual complexity. The first months are spent tracking faces, before we shift our attention to objects, before eventually linking objects with sounds. It is the primal sketch, the scaffold of a rich visual category system that supports everything from spatial reasoning to literacy.

Today, perhaps for the first time in our species’ history, this ancient sequence is unfolding in a very different environment.

What Happens When Early Input Shifts

A study in Nature Communications examined adults who had congenital cataracts removed in infancy. The researchers found that a short period of early visual deprivation left lasting effects on the earliest stage of the visual system. The primary visual cortex, which processes basic features such as contrast and orientation, never fully recovered.

Yet higher-level areas in the ventral occipito-temporal cortex (VOTC), which recognize categories like faces and objects, looked surprisingly typical. These regions developed normally through years of later visual experience. This finding confirms what developmental scientists have suspected. Different parts of the visual system have different sensitive periods, windows that close during specific periods of life. Some close very early. Others remain open much longer and depend on years of exposure to the right kinds of patterns.

This is where modern life enters the story.

The New Visual Diet

Screens can appear in an infant’s world as early as 3 months. Sometimes they appear even earlier when a parent is holding a phone while holding the baby. None of this is intentional, but it changes the visual field the infant encounters. The depth of a room becomes a flat foreground. The rich variability of a face becomes a uniform, glowing surface. The dynamic motions of the body are replaced by the linear motion of a scrolling feed.

If the visual diet during early infancy shifts in this way, the infant’s developing systems may face a question evolution never prepared for them. Should they pay attention to the complex three-dimensional patterns of human faces, or the highly simplified patterns of a two-dimensional screen.

Neuroemergence sheds light on the stakes. Tiny shifts in early input can reorganize the system that later supports more complex abilities. Early face tracking gives infants a basic pattern-recognition system that later supports object recognition and even reading. Our visual system becomes specialized through repeated interactions with the kinds of patterns our environment supplies.

If the environment changes, the pattern changes.

Two Stages of Possible Disruption

There are two windows where screens may alter this developmental cascade.

First, during the first year of life, infants rely on parents’ faces to drive CONSPEC and the transition to cortical face areas. If parental attention is frequently absorbed by a phone, there are simply fewer moments of mutual gaze. These moments are like the drip-drop of rainfall. Small imperceptible drops accumulate and over time can shape the landscape.

Second, between ages 1 and 5, children begin looking at screens themselves. This period overlaps with the sensitive window for tuning higher-level visual areas, including those that recognize faces, objects, and early written symbols. If an infant sees thousands of faces during play and early conversation, the system learns the complexity of real faces. If much of the input is flattened and stylized, the system may specialize in ways that reflect that difference.

None of this means children will lose the ability to recognize faces. The concern is subtler. The brain may become less attuned to the microexpressions and social cues that require extended face-to-face exposure. In Mastery, I noted that expertise grows through repeated exposure to complex patterns. We can become experts in faces, in dogs, in cars, in birds, or even in alien “Greebles” if we look at them often enough. The developing brain becomes expert in whatever it sees most.

The question is what the modern infant sees most.

Why This Matters for Later Cognition

Face processing does not stand alone. It is part of the broader network that supports social cognition, spatial reasoning, and the coordination of vision and movement. Reading itself grows in an area that sits beside the fusiform face area. Small changes in early face tracking can shift how later systems are built.

If we alter the early stream, the downstream structures may follow suit. This could influence how easily children read eye gaze, detect subtle emotional cues, or sustain attention on the physical world. These changes may not show up on standard tests. They may appear instead in the ease of conversation, the depth of social understanding, or the intuitive grasp of space.

What We Can Do

This is not a call to eliminate screens. It is a call to align our habits with what we know about development. The first three years are a time when the visual system is planting the seeds of our visual world. During this period, face-to-face interaction provides the natural signals the brain expects.

Parents do not need to be perfect. They simply need to know that these early interactions matter. A few more moments of shared gaze at mealtime or during play can reinforce the patterns the infant’s brain is designed to learn from.

We are living through a developmental experiment that no previous generation has experienced. Children will adapt. They always do. The question is how. Understanding this now will help us guide the next generation toward a world where technology enhances development rather than flattening it.

For a narrative exploration of these themes, see my essay “The Cancer Stick of the Soul” on Substack.

References

Johnson, M. H., Senju, A., & Tomalski, P. (2015). The two-process theory of face processing: Modifications based on two decades of data from infants and adults. Neuroscience & Biobehavioral Reviews, 50, 169-179. https://doi.org/10.1016/j.neubiorev.2014.10.009

Morton, J., & Johnson, M. H. (1991). CONSPEC and CONLERN: A Two-Process Theory of Infant Face Recognition. Psychological Review, 98, 164-181. https://doi.org/10.1037/0033-295X.98.2.164