RESEARCH_FILE
The Comparison Engine
The comparison engine is the automatic cognitive process — first formalized by Festinger in 1954 — that scans the social environment for reference points and ranks you against them. It runs continuously, not just when you consciously choose to compare. On social media, the algorithm feeds this engine a curated stream of highlight-reel targets, particularly in domains of ability where comparisons are both more automatic and more painful than opinion comparisons.
SEE THE PRACTICE
Turn a thought this research explains into one clear move.
THE THOUGHT
“They're all smarter than me”
YOUR RECORDED RESPONSE
“I’m comparing my inside to their outside. That is not a fair test.”
ONE PRIVATE MOVE
Open a blank note and write three columns: “what I saw,” “what I assumed,” and “what I actually know.” Fill in one recent meeting, email thread, or conversation in under ten minutes, then close the note without editing it.
How it sounds in your head
The inner script: 'Why do I always feel worse after 20 minutes of scrolling, even when nothing bad happened?' The research answer: the comparison engine ran the whole time, benchmarking you against a feed of curated peaks. You didn't choose to compare — the environment chose for you.
TEST_YOURSELF · How well do you know this science?
01 According to Festinger, what type of standards do people most prefer when the comparison engine runs?
Festinger proposed that people prefer to compare with similar others — not those dramatically above or below — when objective standards are absent. Social media breaks this by surfacing dissimilar, highly curated peaks as the default comparison pool. source ↗
02 Which type of comparison did the 2023 Instagram study find more harmful to well-being?
Ability comparisons (d = 0.32) produced larger well-being decrements than opinion comparisons (d = 0.19) in the 2023 Instagram experiment. This aligns with Festinger's original insight that ability evaluations are more driving than opinion evaluations — with platform algorithms now optimizing for exactly the ability-display content that makes the engine run hottest. source ↗
03 Tesser's SEM model predicts comparison pain is sharpest when which conditions are met?
Tesser's SEM model predicts maximum threat when a close other (not a stranger) outperforms you in a domain that is self-relevant (not peripheral). This is why an Instagram friend who excels at your own craft stings more than a famous stranger who excels at the same thing. source ↗