Why eye contact is so hard on video calls (and how to fix it)

Video calls fight one of the most basic things human beings do: look at each other. The camera sits above your screen. The person you're talking to sits on your screen. Your eyes can be on one or the other, but not both. The result is that real eye contact on video is impossible by default, and the consequences run deeper than most people realize.

Desktop setup with an Airtime teleprompter mounted above the monitor, restoring eye contact on a video call

The fix is a setup change in four steps

If you spend hours every day talking to other humans through a screen, that's hours every day where the medium is fighting you. Here's why, and what you can do about it.

  1. Step 1

    A teleprompter, like the one in the Airtime + Elgato Conferencing Kit, restores eye contact by putting the person you're talking to in line with the camera.

  2. Step 2

    Airtime Creator does the same for presentations and recorded video, with speaker notes that sit near the lens.

  3. Step 3

    Airtime Stacks gives everyone on the call a named cursor on a shared document, so people can see where you're looking when you're looking off-screen.

  4. Step 4

    Airtime Camera is the always-on companion app for everything else—lighting, framing, looks.

Three teal semicircles decreasing in size from left to right on a light blue square.

About Airtime

We make essential tools for video at work

JUN 1, 2026 • By Jeremy Brand Yuan

Why humans are built for eye contact

Eye contact and directional gaze aren't decorative. They're foundational to how humans communicate.

There's even a biological argument for it: humans have unusually visible sclera—the whites of our eyes are larger and more exposed than in any other species, including our closest primate relatives. We didn't evolve that way to look good. We evolved that way so other humans could read our gaze with precision. (Communication researcher Rebekah Maggor walks through the science in detail in this video. Most of the research underpinning this piece comes from her work.) Eye contact, what vision scientists call mutual gaze, is one of our top-priority social signals. It's how we gauge whether someone is interested in what we're saying. It's how we establish presence in a conversation. It's how we build rapport.

Directional gaze is the other half. It's our ability to use our eyes as pointers—to look at a document or a person and have everyone in the room understand that's what we're focused on. It works because we share a physical space. We agree on our orientation relative to each other. When you glance at the door, the other person looks at the door. When you focus on a document you're reviewing together, their attention follows. Shared attention is the substrate of collaborative thinking, and it depends on directional gaze to function.

Neither of these signals survives the trip onto a video call intact.

Video calls kinda break our brains

Most people assume video is just a degraded version of in-person communication—same signals, lower fidelity. That's not quite right. Video doesn't only deny crucial visual signals. It sends misleading ones.

The camera offset is the obvious problem. The camera sits above your screen. The person you're talking to sits on your screen. Your eyes can be on one or the other, not both. Every time you read someone's expression, you're not making eye contact. Every time you look at the lens, you can't read theirs.

There's a second, subtler problem. When you glance off-screen on a video call, your conversation partner has no idea what you're looking at. Your slides? Your email? Another person in the room? They can't tell, because they don't share your physical orientation. So every off-camera glance reads as inattention by default. It's not that they can't see what you're looking at—it's that they unconsciously interpret your gaze as wandering, even when you're paying close attention to something directly relevant.

This is why video calls feel so exhausting. It's not the screen time. It's that your brain is working hard to constantly correct what you're seeing against what you imagine to be reality. The result is that you end up feeling exhausted, detached, and ill at ease—even when the conversation itself went fine.

The stakes get higher the more important the conversation. A lawyer on a video call with a client, a psychologist seeing a patient, a teacher working one-on-one with a student, a salesperson trying to build rapport with a prospect, a candidate being interviewed for a job they care about—these are all situations where eye contact and shared attention carry real weight. The default video setup quietly undermines all of them.

What works: the four-step fix to better eye contact on video

Once you understand what's broken, the fix is straightforward. Four steps, in roughly this order. The first three handle specific situations where eye contact and shared attention matter most; the fourth is the always-on layer that supports all of them.

Step 1: For eye contact, a teleprompter

The mechanical problem is that the camera and the person you're talking to are in different places. The mechanical fix is to put them in the same place.

A teleprompter does exactly this. A half-mirror reflects your screen back at you, with the camera positioned directly behind the glass. You see the person you're talking to, and you look at the lens at the same time. The geometry that made eye contact impossible by default is no longer in your way.

This isn't new technology. News anchors and presidents have used teleprompters for decades, which is why most people still think of them as bulky studio equipment for reading scripts. What's new is a generation of small, affordable desktop versions designed to sit on a regular desk in front of a regular laptop. The Airtime + Elgato Conferencing Kit bundles the Elgato Prompter with the rest of the hardware we'd put on a serious video setup. (Yes, we had a hand in this—Airtime and Elgato co-developed the Kit together.)

You don't need a script to use a prompter. You can leave the screen blank, or pull up notes, or just use it to align the lens and the person's face. The point isn't the script. The point is the geometry.

Step 2: For presentations and recorded video, Speaker Notes near the camera

The teleprompter solves the live-conversation case. The other place the eye-line problem shows up is when you're presenting—giving a webinar, recording a tutorial, walking through slides for a group.

Airtime Creator handles this case with a Speaker Notes view that you can position near the top of your screen, close to the camera. Your slides, talking points, and presentation controls all live in the same focused view, and that view sits directly under the lens. The same underlying principle as the prompter: move what you need to look at into the path between your eyes and the camera.

Not as complete a fix as a hardware prompter—you're still looking slightly below the lens rather than directly through it. But meaningfully better than the default, which is notes at the bottom of the screen, guaranteeing your gaze is downward for most of the presentation.

Step 3: For collaborative viewing, shared cursors instead of screen sharing

Eye contact is one half of the visual-signal problem on video calls. The other half is shared attention: when you and the people you're with need to look at the same thing together. In person, this happens through directional gaze—you glance at the part of the document you're discussing, and everyone's attention follows. On a video call, the audience can't tell where you're looking, so every off-camera glance reads as inattention by default.

The usual workaround is screen sharing, which solves the problem by making one person the driver and everyone else a passenger. Airtime Stacks takes a different approach. Everyone on the call gets a named cursor on whatever document you're reviewing together. When you're looking at the third paragraph, everyone else can see your cursor on the third paragraph. When someone points at a figure, you see their cursor on the figure.

The cognitive load of "I can't tell what they're looking at" disappears, because now you can. And unlike screen sharing, anyone can scan ahead, zoom in on a detail, or point at the part they want to discuss—while everyone else sees exactly what they're doing.

Step 4: For everything else, Airtime Camera as the always-on layer

Airtime Camera addresses a different problem from the rest of this setup. It doesn't fix eye contact or directional gaze. What it does is improve how you show up on camera generally—lighting compensation for bad rooms, framing that keeps you centered when you shift in your chair, looks that make you presentable when you've joined a call before you've had coffee.

We mention it here because it belongs in the same setup. Anytime your camera is on, Airtime Camera runs underneath whatever video tool you're using, doing the work that the platforms don't do themselves. Eye contact and shared attention are the communication signals; lighting and framing are the conditions under which those signals get transmitted. Both matter.

The path not taken

There are other approaches to this problem worth knowing about. Some companies are building expensive multi-camera video conferencing stations that reconstruct your image from several angles. Others are replacing the camera entirely with AI-rendered avatars. Both miss what we think is the central point.

The signals humans use to communicate are real, biological, and already present. They've been refined by millions of years of evolution. The technology shouldn't replace them or reconstruct them—it should get out of their way. The setup we're describing is grounded in real human signals rather than simulated presence, and it's available at consumer prices today rather than at enterprise scale years from now.

Frequently asked questions

The camera sits above your screen and the person you're talking to sits on it. You can look at one or the other, not both. That's the camera offset, and it makes real eye contact on a default video setup impossible by design—not a willpower problem.

Only one-way, and only briefly. If you stare at the lens, the other person sees your eyes but you can no longer see their face or read their expressions. The only setup that fixes both at once is a teleprompter that puts the screen and the lens in the same line of sight.

A half-mirror reflects your screen back at you with the camera positioned directly behind the glass. You see the person you're talking to and look into the lens at the same time. Works with Zoom, Teams, Meet, Webex, and any other app—the prompter is hardware, not software.

Directional gaze is your ability to use your eyes as pointers—looking at a document or person so everyone else knows that's where your attention is. It depends on shared physical space, which video calls don't provide, so off-camera glances read as inattention even when you're focused.

Screen sharing gives one person's screen to everyone else, so only one person controls the view. A shared cursor on a shared document—what Airtime Stacks does—gives everyone the document and lets each person point, scroll, and zoom independently while everyone sees where the others are looking.

It's not the screen time. It's that your brain is working hard to correct what you're seeing—broken eye contact, missing directional gaze, partial faces—against what you imagine to be reality. Restoring eye contact and shared attention cuts the cognitive overhead.

It substitutes a reconstructed image for the real one—it doesn't restore the actual signals other people read from your eyes. Real eye contact and shared attention are biological signals already present; the better fix is to remove what's blocking them, not replace them with a simulation.

No—but if you spend hours a day on video, especially for conversations that matter (sales, interviews, therapy, teaching), a prompter pays off fast. For ad-hoc calls, Airtime Creator's Speaker Notes near the lens and Airtime Stacks for shared attention cover most use cases without dedicated hardware.

Yes. Airtime Camera registers as a virtual camera on your system, so it works with any video app that lets you pick a camera input—which is all of them. Stacks and Creator are standalone apps you can run alongside whatever video tool you're already using.

Better video calls, starting on the next one

There are roughly 1.25 billion knowledge workers worldwide, many of whom spend hours every day on video. The aggregate cost of broken eye contact and lost shared attention is enormous. The aggregate gain from fixing it, even partially, is correspondingly large.

The good news is that you don't have to wait for the medium to get better. The setup change is small. The prompter sits on your desk. The software runs in the background. The next call you take with a Stacks document open, with Speaker Notes positioned near the lens, or with the prompter aligning the camera and the face—that next call will be different from the one before it. People will notice, even if they can't quite articulate why.

We live in an age of digital saturation and AI-mediated everything. The chance to communicate directly with another person, in real time, is more valuable than it's been in a long time. When you get that chance, it's worth a small investment in setup to do it well.

Three turquoise curved shapes decreasing in size from left to right, suggesting a sound wave, on a light blue gradient background.

Ready to look each other in the eye? Try Airtime today.

Restore real eye contact and shared attention to every call. Works with Zoom, Meet, Teams, Webex, and any other app that lets you pick your camera—or take it further with the Conferencing Kit, a complete hardware + software solution, designed and tested together.