June 10, 2026
There's an idea that's been buzzing around my head for the past few months. It's the most "research as mesearch" thing that I might have as a hot take; Zoom might be destroying the field of cognitive development.
Let me explain...
In the Fall of 2019, I won a grant to do work on reach tracking in young children. For a few years, Chris Erb, a graduate student working in my lab, had helped me adapted this technology to study inhibitory control in children. It was a cool idea - basically, we were tired of studies that simply correlated inhibitory control measures with other (more interesting) measure of cognitive development. For example, there's a lot of research in theory of mind that suggests there are correlations between different kinds of false belief tasks and different measures of inhibitory control, particularly inhibitory control that involves suppressing a prepotent response in favor of a more rule based on (my favorite being the "whisper" task - which is to show children a really exciting fantasy character, like Princess Elsa, but having them whisper the character's name. Try this with a 3-year-old.
The grant involved us designing a set of studies to measure children's reaching as they responded to questions. We built a tracker display and connected electromagnetic sensors and built the stimuli, and piloted a study on a few adults. And then the pandemic hit.
This kind of study was designed for in person testing. You have to hook up the sensor to the child's finger and there is special technology that tracks the sensors position in space over time, and there's no way to do this kind of work remotely. We were all set to go in February 2020. And then, COVID. Nothing. No in person data collection. For two years.
So, like everyone else, the lab pivoted. We were able to write up some existing data we had already collected. I wrote a long review paper on pretend play (which includes the all-time favorite experiment I've ever done, which I'll have to write about in another entry), and a short review paper on children's STEM engagement. I conducted a meta-analysis on the false belief task (in truth, that one was already underway, but I finished it during that time). I also wrote a book. But in the lab, we pivoted to collecting other data. My postdoc created an online blicket detector simulation (I still have some objections to using this paradigm, which I'll get to). And we ran some studies on children's fairness. So, you know, productive. And as the world recovered, we eventually did the reach tracking work - looking at the role of inhibitory control in children's fairness norms, social learning, and performance on a false belief measure. I even wrote a review paper on it that's in press in Psych Review.
But at the center of the pivot was Zoom. Zoom studies allowed researchers in cognitive development to study children without a lab. As long as we had a way to communicate with parents for them to sign up, and experiments that could be administered remotely, there was the possibility of collecting data. And just about every lab did this. And many labs never stopped. Many labs in cognitive development never went back to testing children in person.
There's some real advantages about Zoom studies. Researchers don't have to do much - they can advertise remotely and rely on a calendar program to schedule appointments, so they just have to be available to test. The start up costs are minimal as is the time commitment. Parents often like participating in research where they don't have to leave the house. Rescheduling is easy. And, researchers are often able to collect a pretty large sample, often more quickly than working off of a database for in person testing. There's a belief that the samples collected over Zoom are more diverse (I'd love someone to do an analysis of this to confirm this hypothesis). And, there are now a set of services that not only allow for moderated Zoom testing, but unmoderated (i.e., asynchronous) data collection - cases where you program up your study, and parents sign up and children are tested without an experimenter even being present. Passive data collection!
None of what I have said here is bad. But here's the "mesearch" part of this. A study like reach tracking - where you have to bring the kid into the lab and measure their behavior in person using specialized stimuil - just isn't possible over Zoom. As a graduate student, I was part of a team that created the "blicket detector." The detector is a box that lights up and plays music when certain objects are placed on it. There's nothing special about the objects - it's all controlled by the experimenter (usually via a switch under the table, like a magic trick). There's something salient about physically placing objects on the machine and seeing it light up. There's something even more salient to a child about them placing objects on the machine themselves and observing the effect - in fact, that's even some studies that suggest that children interpret the efficacy of their own actions differently than the observation of others' actions.
That just doesn't seem possible over Zoom. Yes, I know that people have built virtual blicket detectors and used them over Zoom (hell, my lab did this). But something feels wrong about it. We as a field don't know if it's the same. My intuition is that it is not.
And I'm not just talking about blicket detector or reach tracking (i.e., the methods that I use). Zoom is amenable to certain methods - telling children stories and asking them questions or engaging with them in direct interview.
There's a limit of what is and is not possible over Zoom. Anything that involves manipulation is less possible over Zoom.