Eerie Coincidences Aren't That Eerie

Show Notes

Is your phone listening to you? You mention something random, something you’d never normally talk about, and within days there’s an ad for it. I used to be sure that was coincidence. Some companies have since tried. But I still think most of it is you.

Rather read? I also made this episode into an article: https://btr.mt/analects/on-coincidence-lecture

And the piece it’s based on: https://btr.mt/analects/on-coincidence.

Stuff I mentioned:

• Eerie coincidences aren’t that eerie—the article this episode grew out of
• 404 Media (2024). Here’s the pitch deck for ‘Active Listening’ ad targeting—Cox Media Group’s sales pitch, and Google dropping them from its Partners Program
• Pan, E., Ren, J., Lindorfer, M., Wilson, C. & Choffnes, D. (2018). Panoptispy: Characterizing audio and video exfiltration from Android applications. Proceedings on Privacy Enhancing Technologies, 2018(4), 33–50—the Northeastern study of 17,000 apps; there’s a plain-English summary from Northeastern
• GrapheneOS—the privacy-hardened Android I use
• Google. Delete, switch, or add users—Android’s user profiles, which you can use to sandbox apps
• How to never get mad again (kind of)—the interruption theory of emotion
• Emotion and the mind—on predictive brains and cognitive misers (the cheapskates)
• Barrett, L. F. (2017). The theory of constructed emotion: an active inference account of interoception and categorization. Social Cognitive and Affective Neuroscience, 12(1), 1–23.
• Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127–138.
• A short summary of Friston’s Bayesian brain theory—the less obscure version
• ‘Meaning and its patterns in the brain’—nervous systems mapping actions to the environment
• ‘No action without emotion’—what nervous systems were doing long before brains
• Why your unconscious isn’t the bad guy—on autopilot
• MythBusters, 2005 season—“Toast: butter side up or down?”
• Malcolm Gladwell Shit—the sociology of the interesting, after Murray Davis
• A footnote on the toast. The folk belief turns out to be right, for physical reasons. Robert Matthews showed that toast sliding off a table doesn’t have the height to finish turning over, so it tends to land butter side down—the study won the 1996 Ig Nobel Prize for physics. Tossed in the air, it’s about 50/50. Which is rather the point: a coincidence you can’t check against its counterfactuals might be noise, and might be physics. Matthews, R. A. J. (1995). Tumbling toast, Murphy’s Law and the fundamental constants. European Journal of Physics, 16(4), 172–176.
• the btrmt. lectures podcast

Related stuff:

• “Purple Doesn’t Exist”—the brain inventing what it can’t account for
• ‘The downsides of the prediction machine’

What is btrmt. lectures?

A brain scientist talking about (better) patterns of thought, of feeling, and of action. One pattern, one podcast—you see if it works for you. The btrmt. lectures, with Dr Dorian Minors. (btrmt.—said "betterment.")

Welcome to the Betterment Lectures. My name is Dr Dorian Minors, and
if there's one thing I've learnt as a brain scientist, it's that
there's no instruction manual for this device in our head. But there
are patterns. Patterns of thought, patterns of feeling, patterns of
action, because that's the brain's job: creating the patterns that
gracefully handle the predictable shapes of everyday life. So let me
teach you about them. One pattern, one podcast, and you see if it
works for you.

Today I want to do a somewhat shorter lecture than previous ones.
I've been getting a little out of hand with the recording time, and
the editing time after it, so I'm going to practise something a
little smaller. Hopefully interesting, though. Something I was
recently prompted to revisit: the idea of our phones listening to
us. And more generally, something I've [written about
before](https://btr.mt/analects/on-coincidence)—the idea of an
eerie coincidence.

So that's today's little lecture. Eerie coincidences. How eerie are
they?

Is your phone listening to you?

I think a lot of people are worried that our devices are listening
to us. At least among the people I hang out with—and maybe that
says more about me than about the phenomenon. But I think a fairly
common opinion, or suspicion at least, is that social media apps and
other apps on our phones are picking up our conversations to sell us
ads.

And I think we think this because it's pretty surprising how often
an advert appears on Amazon, or in the Instagram feed, that reflects
a recent conversation. I'll speak about something completely random,
or something I think is random, something I would never normally
talk about. And lo and behold, within a couple of days, I'm shown an
ad for precisely that random thing.

People try to explain this away by saying these algorithms are
fantastic at learning about you. So fantastic, in fact, that they
predict the things you want and the things you talk about. It's not
that they're listening to you. It's that they're predicting that not
only do you want this stuff, you're likely to talk about it.

What's changed since I wrote the article

When I first wrote an article on eerie coincidences, from what I
could tell, this basically wasn't happening. Now my perspective has
changed a little. I was essentially sceptical of the idea, but we've
since seen examples of companies doing this, and maybe I shouldn't
have been so laissez-faire about it. Of course these companies would
if they could, right?

I'll link to the particular article I'm thinking of, from [404
Media](https://www.404media.co/heres-the-pitch-deck-for-active-listening-ad-targeting/),
which reported on a media group that advertised "active listening"
ad targeting. As a consequence, they were removed from the Google
Partner Program.

And the thing that brought it to my attention at all: somebody
shared with me [a paper from
Northeastern](https://doi.org/10.1515/popets-2018-0030) that looked
at something like 17,000 Android apps. They didn't find any evidence
that these apps were secretly enabling the microphone to send audio.
But they did find apps sending screenshots and screen recordings to
third parties. So this is definitely happening.

But the reason I was sceptical is something I still want to talk
about. And that is that eerie coincidences are probably just
coincidences that the mind makes seem more important than they
actually are. That's what I want to talk about today. Some eerie
coincidences are suspicious, but probably most are just you getting
tricked by your perceptions. So let me explain how.

A paranoid person's phone

I'm a pretty paranoid person, I would say. To give you an example, I
use an operating system called GrapheneOS
on my Android phone. I do that because I'm worried about security
and privacy—though I don't really know what that means. So I
thought I'd go through the fairly difficult rigmarole of getting it
installed on my phone, rather than hand over all my data to Google
on the stock operating system.

Or, as another example: I don't know if you know this, but on your
Android phone you can have [different
users](https://support.google.com/android/answer/2865483), like
different users on a computer, as if multiple people were using the
same phone. And you can use this feature to sandbox apps. So I have
my Instagram in a separate user profile, to keep it away from the
other apps on my phone. Again, not because I'm worried about
anything specific. Just because I'm generally paranoid about stuff
spying on me.

And I know this is pretty weird, because I have a mate who works at
Google, and he saw me doing this one time. He was like, "I'd love to
talk to you, because I was just having a conversation with another
Google colleague about who actually uses user profiles on the
phone." So, pretty paranoid, I think.

And even then, I will still notice oddly specific ads that seem
absurdly relevant to recent conversations, and to nothing else in my
behaviour that I can detect. That's the thing that really made me
think they're probably not listening to what I'm saying. Probably
this is something else. Now, again, we have evidence that some
people are trying to listen, so this isn't always true. But let me
give you examples of other eerie coincidences that probably aren't
that eerie.

We only notice the mistakes

When I mark student work—essays, or the students I'm
supervising—I will almost always only comment on the things that
need correcting. Much to my students' dismay, if what they've
written is good, I almost never stop to make a note of it. It's such
a problem, I think, for students who expect more emotional
intelligence from their supervisors, that I have to say it up front,
and in emails when I send feedback: a lack of commentary means it's
good.

But I don't think this is unusual. I think you probably do this too,
right? Maybe you're reading a book, or watching a TV show, and
you're completely absorbed in the thing until you notice a mistake.
A typo in the book, or the captions on the TV show are wrong or
weird. And all of a sudden you're jolted back into yourself. You're
no longer absorbed in the story. You're paying attention to the
error.

Interruptions and the predicting brain

I think this experience is on the same continuum as the "is my phone
listening to me?" experience. Because it's a feature of our brain,
our nervous system, to come alive—to come especially active—when
it notices interruptions.

The [interruption theory of
emotion](https://btr.mt/analects/interruption-theory-of-emotion-mandler)
is one of my favourite theories in psychology, and I'll link an
article in the show notes for more on it. The core idea is that
interruptions to our experience—interruptions to our
expectations—are the thing that make us stop, pay attention, and
have emotions in response to the world.

Since that theory in the '80s, a lot of brain science has found
itself circling a similar concept. The most recent instantiation of
it, if you're keen to Google it—and I'll give some links as
well—is the [predictive
coding](https://btr.mt/analects/emotion-and-the-mind) stuff from,
say, Lisa Feldman Barrett, or
the active inference stuff from the notoriously obscure writings of
[Karl
Friston](https://btr.mt/analects/30082-a-short-summary-of-fristons-baysian-brain-theory).

Fundamentally, though, our nervous systems appear to have developed
in large part to [map actions to the
environment](https://btr.mt/analects/making-meaning-in-the-brain).
To map the statistical structure of the world and your behaviour in
it, and make sure that whatever you perceive automatically produces
adaptive actions in response. And this function of the nervous
system has been around since [long before brains were a
thing](https://btr.mt/analects/no-action-without-emotion). Before we
had a central nervous system, with a brain on top of the spinal
cord, we just had the peripheral nervous system. The basic nervous
web.

Cheapskates on autopilot

But of course, it doesn't make sense to do this mapping of
perception to action online, all the time—every moment of the day,
your neural system actively working to connect your perceptions to
the best possible responses. Usually the best possible responses are
highly prescribed, over-determined, and the body can manage them
passively.

What I mean is that our environments aren't really very varied. They
can be very complex, and human behaviour is extraordinary in many
ways. But there's a reason so much of our conversation around growth
and wellness concentrates on routines and habits. The environment
might be very complex, but we've spent our whole lives working out
the best paths through it. And the vast majority of what we do is
done on autopilot.

We're cheapskates.
We encode the best mappings of perception to action, and then we let
them run in the background until something happens that makes us
stop and go, "Oh, we need to update this map. Something's wrong
here."

So this is why I won't make a note when my students are writing good
stuff. I'm absorbed in what they're doing. I'm not being
interrupted; it's a coherent argument. Until some strange sentence
or illogical statement jolts me out of that experience, I'm not
going to notice. And it's the same with you and your book or your TV
show. It's flowing smoothly through the brain, meeting all your
expectations about how books should go, or the captions on a TV
show, until you hit an error or some event violates your
expectations. Then you're jolted out. You're interrupted. You wake
up and pay attention in a different way.

Now, all of that might seem a tangent from what started off being
about whether our phones are listening to us. But I think it's
contiguous with that experience, because this interruptive feature
of the brain tells us something. If the world is aligned with our
expectations, we shouldn't notice. When things are expected, we
should essentially ignore them—not be interrupted. We should only
really notice things when they interrupt our experience.

Peanut butter toast

So let me give you an example to bring us back to the point. I love
peanut butter toast. I eat heaps of it. And if it weren't for a
twenty-year-old
MythBusters)
episode that still lives in my head to this day, I would have
thought that whenever I drop my peanut butter toast, it lands peanut
butter side down.

If you test this, what you'll find is that it's just as likely to
fall peanut butter side up as down. And the reason you probably
think one happens more than the other is: why would you notice when
it falls peanut butter side up? There's nothing wrong. You just
start eating it.

The only reason I think it falls peanut butter side down is that I
only get interrupted when it lands upside down. If it's peanut
butter side up, I can just pick it up and start eating. I don't need
to get the cloth and wipe the floor.

The ads you never notice

So basically, I think this is more or less what's happening with the
listening phone thing. We get inundated with ads all day, every day,
and you never notice when the ads are random and unrelated to
conversations you've had recently. You're good at ignoring useless
ads, useless information. You only notice the really strange
ones—the ones that are just generally peculiar, or the ad that's
strangely topical given what you've been talking about with your
mates. But the proportion of ads you notice, against the ads you
can't remember, is probably super, super tiny.

That's my speculation, anyway. Maybe somebody could do an undergrad
research project on it. But probably most unusual coincidences are
this.

So let me close out with a couple of implications, and then we'll
wrap up, and hopefully this will end up being a super short podcast.

Interesting isn't the same as true

I could just leave the lecture there. Probably most unusual, eerie,
suspicious coincidences aren't unusual, eerie or suspicious. They're
just a matter of you not noticing the counterfactuals. You're not
comparing the relative likelihood, because you can't remember all
the times it went normally. Your entire nervous architecture is
tuned to detect and recognise interruptions to your expectations,
not the things that are contiguous with them.

And I will essentially leave it there. But I will point out that I
very often talk about what's known as the [sociology of the
interesting](https://btr.mt/analects/education-is-entertainment),
and I'll see if I can link to a couple of places where I talk about
it more. Essentially the idea is that whether something is true
doesn't seem to matter much for whether we find it interesting. For
us to find something interesting, it needs to violate our weakly
held beliefs. We need something like a hot take on something we
don't care very much about.

And that's related to what I'm talking about here. If I tell you
something you already know, you're just going to nod along and
forget about it, for the same reason you don't notice the things
that don't violate your expectations. And if I tell you something
that violates your strongly held beliefs, you're just going to
ignore me. Possibly because it's wrong, but also because there's too
much cognitive and emotional infrastructure invested in our strongly
held beliefs to let a piece of evidence overturn them.

I think the reason I keep talking about the sociology of the
interesting is that it's notable just how much of the internet is
this kind of content. It's what I try to do in my podcasts, and in
my articles at times, to generate interest. It's a pattern that
underlies a lot of the media we find entertaining. Spooky
coincidences are basically our preferred form of content. And I
think that's a problem, because it doesn't need to bear any
relationship to the truth.

So when I first wrote this article on eerie coincidences, the idea
was to get you to question whether an eerie coincidence was really
eerie. And I guess the rule translates to content too. If something
violates your weakly held beliefs and you find it interesting, maybe
that's a signal to stop and think. Maybe this isn't that
interesting. Maybe my nervous system is just tricking me into
thinking that it is.

And that, I think, is where I'm going to leave it today. Until next
time.