What I Found When I Actually Tested My Hunch About Awe

2 days ago
7 min read

Updated: 1 day ago

The 0.04 I couldn't stop staring at. — *The 0.04 I couldn't stop staring at*.

I remember the exact moment I realised my hypothesis was wrong. It was a Friday. I had the correlation matrix open on one screen and a half-finished cup of coffee going cold next to my laptop. I scrolled to the row I was most excited about - the one I'd been thinking about for months, and the number staring back at me was 0.04.

Zero point zero four.

That's not a small effect. That's not even a real effect. That's noise.

I sat there for a while. I refreshed the analysis. I checked the variable. I checked it again. The number didn't move.

Earlier this year, I wrote a post about looking up. About sunsets. About the strange relief of forgetting yourself for a moment. I made a case, backed by other people's research and one very persuasive Dacher Keltner book, that maybe the cure for our self-focus epidemic isn't more self-focus done better. Maybe it's awe.

Then I did what researchers are supposed to do. I tested it.

This is what happened.

The Setup

If you've read my previous post, you know the hunch: appearance anxiety, depression, and that constant low-grade hum of am I okay? are all driven by being trapped inside your own head. And awe, where those moments where the self goes quiet, might be the way out.

It's a beautiful theory. It's also a theory I had to be honest about. I hadn't tested it in my population (i.e., Malaysian emerging adults) with my measures. So earlier this year, with help from my brilliant research interns, we ran a pilot study with 80 university students, ages 18 to 25. We asked them about their appearance worries (Veale et al., 2014), their mood over the past week, how often they got stuck in self-focused thinking (Kiropoulos & Klimidis, 2006), and how prone they were to ruminating (Treynor et al., 2003). And then, the part I was most excited about, we asked them to recall a recent experience of awe and rate what it was like, using the newly developed Awe-SM (Kannis-Dymand et al., 2025).

Here's what I found. Some of it confirmed what I already believed. Some of it didn't. And the part that didn't is, I think, the more interesting part.

What Worked: The Self-Focus Trap Is Real

When I look at the numbers, the self-focus story is unmistakable. The more time my participants spent stuck in their own heads: analysing themselves, comparing themselves to others, replaying past moments, the more appearance anxiety they reported. The relationship was strong, statistically robust, and exactly what the existing literature predicts (Mor & Winquist, 2002).

But what surprised me was which kind of self-focus mattered most.

I had assumed (and so does most of the literature) that appearance anxiety would be most strongly tied to public self-focus: the kind where you're worried about how you look to other people. The hair-checking. The mirror-glances. The wondering whether your outfit reads "trying too hard."

It wasn't.

The strongest unique predictor of appearance anxiety in my sample was private self-consciousness - the introspective kind. The constant figuring myself out. The "I am often absorbed by thinking about myself". The "I think about my past experiences over and over". The kind of looking inward that feels productive but mostly just keeps you in a loop.

That distinction matters. It means

appearance anxiety isn't really about appearance, at the deepest level. It's about a relationship with the self - a relationship where you can't stop turning yourself over and inspecting. The mirror is just where the inspection becomes visible.

That's a finding I want to sit with for a while. Because what it's really saying, uncomfortably, is that the people in my study who suffer most about how they look aren't the ones who care most about how they appear to others. They're the ones who can't stop looking at themselves from the inside. Who replayed yesterday's awkward conversation at 2 am. Who catch themselves mid-comparison and then ruminate on having compared themselves at all.

It also fits with what Mor and Winquist's (2002) meta-analysis of 226 studies showed two decades ago: that private self-focus is more strongly tied to depression and generalised anxiety, while public self-focus tracks more with social anxiety. My data adds a small twist: in this sample, private self-focus also reaches into appearance anxiety in a way the literature has not always emphasised.

What Didn't Work: My Theory About Awe

And now we get to the part I have to be honest about.

I expected to find that people who had vivid, absorbing awe experiences would also be less stuck in their own heads. The "small self" theory (Piff et al., 2015; van Elk et al., 2019) predicts that the more you've felt absorbed by something larger than you, the less self-focused you should be in your daily life.

The data said: no. Or, more accurately, the data said: not really, not the way you measured it.

The correlation between people's reported awe absorption and their self-focused attention was essentially zero. The correlation with rumination was tiny and not statistically meaningful. The buffer I'd hoped for, in my pilot data, was not there.

*The red block is the self-focus trap. The crossed-out cells are where awe was supposed to show up. It didn't.*

I want to tell you my first reaction was scientific equanimity. It wasn't.

My first reaction was: oh.

Then a longer pause. Then a quieter, more honest version: oh no.

Because if the awe theory didn't hold, then the post I'd written was wrong, the talk I'd been giving was wrong, and the direction I'd been quietly orienting my career around was at best incomplete.

I closed the laptop. I made another cup of coffee.

I let the sentence: What does this null result actually mean? sit on the table next to me until it stopped feeling like an attack and started feeling like a question.

Sitting With a Result You Didn't Want

Here's what I've come to think.

The way I measured awe was retrospective (i.e, recall a time you felt awe and tell me about it). That's a snapshot of a past moment, not a measure of how often someone experiences awe in their life, and it's certainly not the same as feeling awe right now. Most of the famous "small self" studies, including the ones that inspired me, used state awe: they showed people something awe-inspiring in the lab and measured what happened immediately after (Piff et al., 2015; van Elk et al., 2019). That's a fundamentally different thing from asking someone in a survey to remember.

There's also something culturally interesting buried in here. My sample was almost entirely Malaysian Chinese, a collectivist context. The aspect of awe that did correlate with lower distress in my data wasn't absorption (the attentional kind). It was connectedness: the feeling of being at one with others, with nature, with something larger. That's a quieter finding, and it might be telling me something I wasn't expecting: that in a culture where the self is already more relationally defined (Markus & Kitayama, 1991), what awe offers isn't necessarily escape from the self; it's expansion of the self into connection.

I'm not ready to commit to that interpretation. It's one pilot, eighty people, and a single statistical correlation. But it's the kind of finding that makes me want to keep looking.

What This Means for What Comes Next

I'm not abandoning the awe research. If anything, I'm more committed to it, but in a more careful, more honest way than before.

The next study, which my lab is launching soon. It will be larger, more diverse, and pre-registered, so I commit to my analyses before seeing the data. After that, I'm planning to bring awe into the lab as an experience rather than a memory, which is the only design that can actually test the small-self hypothesis (Chirico et al., 2017). My pilot can't.

So: the theory might still be right. The pilot just couldn't see it through the measure I used. That's a useful thing to learn before spending years building a research programme on a method that wouldn't have worked.

The Lesson I Almost Missed

Here's the part I find most important, and the part that has the least to do with statistics.

In the original post, I wrote: "We're not anxious because we have low self-esteem. We're anxious because we can't get out of our own heads." My data confirmed the first half of that sentence with overwhelming clarity. The second half, that awe is the way out, turns out to be a question I haven't actually answered yet.

And honestly? I think this is what science is supposed to feel like. Not the satisfying click of a hypothesis confirmed, but the harder, slower work of the data partly agreed with me, partly didn't, and now I have to think more carefully.

Pilots exist for exactly this reason. Mine pushed me toward what is, I think, a more interesting question:

What if the part of awe that matters most isn't getting smaller - it's feeling more connected?

I don't know yet. That's the next study.

A Note on Doing This in Public

I've been thinking about why I wanted to write this post when the cleaner thing would be to wait until I had results that fit the story. Two reasons.

The first is that academia, even the open-science version of it, doesn't do a great job of showing the messy middle of research. We publish the polished paper years after the fact and pretend we knew all along. We don't. Or at least, I don't.

The second is that I think the people who fund research, the students who do research with me, and the public whose lives this research is supposed to touch, all of you, deserve to see how it actually goes. Including when my hunches turn out to be partially wrong.

If you're someone who has felt awe, who has chased that feeling, who has ever pulled over to look at a sunset, I haven't lost faith in any of that. I've just learned that measuring it is harder than I thought, and that what we eventually find may be more interesting than what I started looking for.

Earlier this year, I asked you to look up.

Today, I'm asking you to be patient with me while I figure out what looking up actually does.

Thank you for being here for the messy middle.