Cognitive Throttling
As you read this, the U.S. is steering toward military strikes on Iran, with two enormous aircraft carriers moving into possible near the Strait of Hormuz. We are on the brink of war.
Hold that thought — it connects to everything below.
Emotional Contagion
You may have been among the 689,003 people whose social media feeds were deliberately adjusted to intensify emotion. In 2012, Facebook ran an experiment testing how emotionally evocative posts affect behavior. The feed was tuned to heighten the fervor of what they saw, and the results revealed that emotions are contagious. We amplify and reflect parallel sentiments when our friends experience them. If a friend is thrilled, stressed, concerned, furious, we echo them, then pass that mood along.
Users did, in fact, stay on the site longer. But a more consequential finding was off-screen. Neuroscientists from the University of Pittsburgh explained that they watched brainwave patterns and noticed that as emotional mayhem amplifies, the brain skips the frontal lobe (hypofrontality) and cognitive function decreases significantly.
As emotional volatility rises, executive function can drop. Under heightened arousal, the brain relies less on the frontal systems used for restraint and deliberation, and more on fast, reactive circuitry. Anger is a clean example. Sharing anger can make people feel confident, more willing to take risks, and less patient with consequences. Even positive emotions, like the rush when your team wins, can skew judgment, pushing decisions toward impulse over reflection.
So, what looked like an innocent check-in with friends over social media had an offline physical-world impact. These users experienced a dumbing-down. Their exercise in unmindfulness reduced cognitive function.
Research and whistleblower testimony describe an engagement-optimized system, run by a trillion-dollar company, that “repeatedly resolves conflicts between public safety and profit, in favor of profit.” They observed polarization and documented hate speech that led directly to “actual violence that harms and even kills people,” as whistleblower Frances Haugen testified before the Senate. Internal documents show that Facebook realized that degrading self-control was dangerous, but chose the success of their advertising clients over their users’ lives.
This knowledge doesn’t keep me from using Facebook to stay in touch with family and friends. I’ve treasured it as a way to watch my nieces and nephews grow up.
Disinformation as a Business Model
Numerous studies suggest social media increases illogical thinking and makes one prone to confirmation bias. This reminds me of the time my friends and I saw this headline on Facebook, “FBI Agent Suspected In Hillary Email Leaks Found Dead In Apparent Murder-Suicide.” It was around the time when the Anthony Weiner sexting scandal broke. Half a million shared that headline, and 1.6 million clicked through to Jestin Coler’s utterly made-up story. It netted him thousands of dollars in advertising revenue for his nascent disinformation company. He posted it on Nov 5, 2016. Two days before Hillary Clinton lost the election to Donald Trump.
Jestin Coler, a registered Democrat and married father of two, did not produce the article with malicious intent. He was not paid by a political campaign. The piece was created only for monetary gain. Juicy lies maximized his ad-click revenue. His company was “optimizing for results.” When he was asked by a reporter, “What can be done about fake news?” he replied, “Some of this has to fall on the readers themselves. The consumers of content have to be better at identifying this stuff. We have a whole nation of media-illiterate people. Really, there needs to be something done.”
When asked why he only produces fake news aimed at riling up conservatives, he said, “We’ve tried to do similar things for liberals. It has never worked. It never takes off. You’ll get debunked within the first two comments, and then the whole thing just kind of fizzles out.”
Nigel Farage, who supported the Leave Brexit campaign, was sharing the stage with Facebook CEO, Mark Zuckerberg. Farage boldly proclaimed that without misleading social media ads, there was “no way the Leave Brexit vote, the election of Trump, or the Italian election, could ever possibly have happened.” Zuckerberg received this statement with a wide, prideful smile. After all, in less than a decade, he built a social community of such scale as to unintentionally influence three international elections with what, in retrospect, seems like extremely obvious misinformation. Would the Jan 6 insurrection have happened without social media?
The Liability Vacuum
If Jestin Coler had published his articles in a print newspaper, he would have been subject to lawsuits. Posting online, however, is often protected by Section 230 of the 1996 Communications Decency Act. It guarantees that no online media platform is liable, even if articles published on their network are incorrect.
Attention, privacy, democracy, freedom, rights: these are things you often don’t appreciate until they are gone.
Congress almost regulated social media. There was bipartisan support to improve child safety with the Kids Online Safety Act (KOSA). It passed the Senate with a 91-3 vote in 2024 but stalled in the House. Recent congressional acts have focused on amending Section 230, requiring parental consent for minors, and increasing algorithmic transparency. They stall due to heavy lobbying by tech companies.
We need them to create a legal tool chest to draw the privacy line, ensure the digital stratosphere is a humane place, and create globally agreed safeguards for machine intelligences.
Dopamine Cadence
In the case of online and mobile gaming, data collection occurs completely below a consumer’s radar. Games have a hypnotic power over their players, and as Shoshana Zuboff points out, players can be enticed to physically move to a location in the real world for a reward, as in the case of Pokémon Go. This type of deliverable is called “footfall,” and it is lucrative. Niantic Corp received approximately $900,000 per day from just McDonald’s Japan for players visiting to engage in the augmented reality, “Pokémon Go Gym.”
Protecting humans from being influenced by games, is just as challenging as getting them to accept a vaccine that will save their life. Say, for example, you were to add a warning label to alert consumers. Would it be heeded? A few years ago, Target’s Australian division pulled Grand Theft Auto from 300 stores because the game “rewards the players who commit sexual violence and kill women.” The country banned it. Oddly, the game proceeded to sell online to over 100 million customers. The controversy may have even helped it fly into those homes.
An analysis of a player’s gaming history reveals what does and doesn’t motivate them and where their reasoning is vulnerable. Those weaknesses are optimized for monetization. Little about gaming is secure or private. The game is playing you at least as much as you are playing the game.
Kids in their own homes, even young ones curled up beside their parents, are vulnerable to predatory situations. Of course, I am referring to pornographic messages popping up in games. Once read, there’s no way to unread a creepy invitation.
The product isn’t the game, it’s the player
Would you let your child learn about science via an entertaining website that boasts it delivers “the best, not to be found elsewhere, heroic” stories? The Hamas Al-Fateh website has millions of young subscribers. A team at Yale looked into it and uncovered that it was a recruit vehicle run by terrorist. The group buys geolocation info to narrow-cast for specific demographics. They filter for marginalized young women and alienated youth. They then email links to parents as the website presents as a place to learn science and discuss the jihad. The hero or heroine arrives at a playground, school, or park and when they blow themselves up, the player receives points based on bystander deaths. Suicide terrorism is how players win “heroism points” to join the science winners’ group and enjoy the dopamine rush of receiving a “cyber-ceremony” as they move up to the next level. I couldn’t imagine a parent who wouldn’t realize that this was grooming for one-way missions, but, then again, the shockingly violent Grand Theft Auto, sold over a hundred million copies.
No desire for human connection
Gaming has an ominous impact on neural activity. Researchers at Harvard discovered the cadence of dopamine rush to dopamine rush alters the path of brain activity in players. Their neural brain-wave pattern is indistinguishable from that of a long-term drug addict. They also reported that “players” increasingly find that they prefer time with their games over time with their family and friends. When they play, their heart pounds and they say, “that is when they feel most alive.”
Would a warning like the one on the side of cigarette pack work? It could say, this game may result in you no longer desiring human connection.
Aza Raskin, the interface designer who invented “infinite scroll,” later described the engagement model with memorable candor: “It’s as if they’re taking behavioral cocaine and just sprinkling it all over your interface,” a drug that keeps the hand moving even after the mind has gone dim. Reed Hastings, the CEO of Netflix said, “Sleep is my greatest enemy.”
If only these technology companies had the goal of helping humans thrive? Instead, we have a suicide rate that is more than double US homicides, surpassing auto accidents. It is the leading cause of death for young people between the ages of 10 and 34.
The Center for Humane Technology frames social media as humanity’s “first contact with AI: a recommender system choosing what hits our retinas and eardrums, and doing so under incentives that reliably misalign with human flourishing.”
A Fake News Machine
The artificial intelligence called GROVER was created to help humans expose fake news. To do that well, it had to first learn to generate it. Its fake news was so believable that 72% of humans in one cohort thought they were reading authentic New York Times articles. GROVER was built at the Allen Institute for Artificial Intelligence in partnership with the University of Washington to prevent what they describe as the “major societal problem” of “news designed to intentionally deceive.”
They trained it on an enormous 120-gigabyte library of news from 2016 to 2019. Essentially the knowledge “canon” that you, the reader, and I, a human writer, had consumed. It took two machines working in a generative adversarial partnership. One machine gleaned this massive breadth of print knowledge as raw material and taught itself how to mirror it, thereby creating fake articles. The other machine used this same data to become informed enough to rate whether the articles created by its partner machine were convincing.
Grover itself can detect fake from real 92% of the time. Thus 8% of the time, it too is fooled. This idea of using pairs of AI computers in a feedback loop, one to generate and one to rate the output, facilitates rapid improvement, and to a large extent, once set up, they can teach themselves, given they have enough of a source library.
Because of the “surplus” of human data sneakily pulled from us via social media, there was a “large enough source library” to train generative adversarial networks (GANs). Will a tech company use their machine intelligence to trick us with fake news thereby dictating our actions?
Harvesting and Curating
Mark Zuckerberg is in the witness box this week, compelled by a judge who rejected Meta’s argument that his in-person testimony was unnecessary. He is being asked point blank whether Instagram was engineered to hook children and monetize their attention.
Google’s YouTube and Meta Platforms are the two remaining defendants in the case, as TikTok and Snap have settled.
Judge Carolyn B. Kuhl started by warning the courtroom not to use AI smart glasses to record or run facial recognition on the jury. The stakes are high. jury tampering is a concern. This is a bellwether case. It is not about the defendant, it is about social media knowingly manipulating humans and profiting on their vulnerabilities.
This isn’t the first time he has been in the crosshairs of a congressional or judicial proceeding and historically he emerged essentially unscathed. But this time feels different.
Zuckerberg shared that he was allergic to the word addiction and that Instagram already does not allow users under 13. His argument is that Instagram is not intentionally addictive and that scientific evidence does not prove social media causes mental health harm to teens.
He was asked about “beauty” filters, as experts warn they intensify body dysmorphia. Zuckerberg defended keeping them as a matter of “free expression.”
Zuckerberg pointed out that teens account for less than 1% of Instagram revenue.
The plaintiff’s attorney, Mark Lanier, caught the reference to revenue and jumped on it making sure the jury understood that profit is the companies measure. Harvesting attention, inference, behavioral conditioning, and monetizing attention, is just how they get it.
When the frontal systems go quiet, persuasion becomes easier. It is a slippery slope. Let us agree to not be a population nudged toward impatience, overconfidence, grievance, and shortcut thinking.
Today’s post is dedicated to a women’s ski trip. After a weekend of beauty and companionship in Tahoe’s backcountry, on Castle Peak, the fifteen skiers were overwhelmed by an avalanche. They were experienced skiers, but it happened quickly. It is heartbreaking. Nine lives are presumed to have been taken, six miraculously survived.
NEWS…
Once again, President Trump steers us to the brink of war. The excuse, which is surely not the reason, is that our nuclear negotiations with Iran have failed. Two enormous aircraft carriers are moving into position near the Strait of Hormuz in the Persian Gulf.
The United Kingdom blocked us from using their military bases for potential air strikes against Iran, citing that pre-emptive strikes would breach international law; wisely, they also fear being drawn into the conflict. One imagines they may have also had quite enough of President Trump. Making matters worse, Trump, in retaliatory dramaturgy, has gone on social media, saying that he withdraws his support for the UK to cede sovereignty of the Chagos Islands to Mauritius.
Luckily his State Department has been clear with the UK that they fully support the ceding of territory and will hold the line. This seesawing of Trump and his administration is a sign that guardrails restraining Trump are appearing.
It hailed here this morning and I went outside with my dog, Toby, to watch from the front steps. Behind us the usually timid juncos, finches, and chickadees, left the trees and clamored for a spot on the porch. There must have been two dozen trying to avoid being bonked on the head by the hail stones. I was enjoying the commotion they made and stayed very still, lest they spook and flurry away.
Thank you for reading. Thank you for supporting independent news. Though the DOJ is still trying to flood the zone and the ICE violence hasn’t stopped, our brilliant mayors and governors are standing up for our rights. Their eloquent and insightful state of the state and state of the city addresses let us know how hard they are working.
Our legislators have been tormented by Trump and have lived in fear, but something is changing. They are thinking independently. Is it the protests? Is it the deep discussions at Davos and the Munich Security Conference? I am glad our legislators, past administers, and governors attended those conferences. I think what they heard and shared is emboldening them.
See you tomorrow.
-Petra
Another brilliant piece. Thank you for the insight on the whistleblower research, the liability vacuum and the dopamine cadence. So frightening. I really feel for anyone trying to raise a child today.