Chapter 10 Key Takeaways: Social Rewards and the App...

The like button was introduced to Facebook on February 9, 2009. Justin Rosenstein proposed the "awesome button" concept in 2007 as a frictionless way to express appreciation. After approximately two years of internal debate about its psychological effects, it launched. Its spread to Instagram, Twitter, TikTok, and every major social platform has made it the universal grammar of digital social feedback.
Social approval activates genuine neurological reward circuits. The ventral striatum and medial prefrontal cortex — the brain's primary dopaminergic reward structures — respond to social acceptance with the same signals they use for food, warmth, and other primary biological rewards. This is not metaphorical. Social approval is a primary reward, not a secondary or derived one.
Social rejection activates the same circuits as physical pain. Eisenberger and Lieberman's 2003 Science research demonstrated that social rejection activates the dorsal anterior cingulate cortex — the same region activated by physical pain. The colloquial experience that emotional rejection "hurts" reflects a neurological reality. Getting three likes on a post you hoped would do well is not a trivial disappointment; it activates genuine threat circuits.
The evolutionary logic of social reward is millions of years old. For most of human evolutionary history, social acceptance and rejection were survival-relevant: exclusion from the group meant dramatically increased mortality risk. The brain's reward and pain systems reflect this history. Platforms exploited an ancient mechanism, not a modern superficiality.
Adolescents are neurologically maximally sensitive to social reward signals. Research consistently shows that the adolescent brain's social reward circuits — particularly the ventral striatum — are hyperresponsive to social feedback compared to adult brains. Maya at seventeen is in the developmental phase for which the like button's effects are most potent. The platform's core teenage demographic is also its most neurologically vulnerable.
Quantification fundamentally changes social approval dynamics. The shift from qualitative social feedback (you sensed how people responded to you) to quantified feedback (a specific public number) enables direct arithmetic comparison, creates public social verdicts attached to content, and transforms social standing from a contextual perception into a legible score. Festinger's social comparison theory predicts the comparison behaviors that follow.
Leah Pearlman and Justin Rosenstein, core creators of the like button, have publicly expressed concern and regret. Their candor — specifically Pearlman's acknowledgment that she became addicted to the feature she helped build, and Rosenstein's account of limiting his own social media use — provides documented evidence from people with the deepest institutional knowledge of the like button's design intent and actual effects.
Variable social reward is more behaviorally powerful than predictable approval. The uncertainty of how a post will perform — the gamble embedded in every share — creates variable reinforcement dynamics (see Chapter 8) that produce stronger neurological responses and more compulsive engagement than predictable approval would. The like button is not just a delivery mechanism for social reward; its uncertainty is itself part of the reward mechanism.
Instagram is a social comparison machine with three amplifying factors. The combination of (a) universal public like counts enabling arithmetic comparison, (b) algorithmic curation that presents high-performing content as the baseline experience, and (c) comparison access to thousands of accounts rather than dozens of social contacts creates a comparison environment categorically more intense than ordinary social life.
Internal Meta research documented that Instagram made body image issues worse for 32 percent of teenage girls. This finding — from a 2019/2020 internal study reported by The Wall Street Journal in 2021 — demonstrates that the platform had substantial internal evidence of harm before the 2019 hidden like count experiment. The company continued operating with minimal changes while possessing this knowledge.
Comments carry higher reward value than likes because they signal greater effort. The effort asymmetry — a like requires one tap, a comment requires thought and typing — makes comments stronger evidence of genuine appreciation. Research confirms stronger neural reward responses to comments. Platform notification and display systems reflect this hierarchy, with comment notifications typically receiving higher priority and prominence.
The approval economy shapes what users choose to share. Maya's content self-censorship — choosing not to post things she liked because she feared the approval verdict — illustrates a mechanism by which the approval economy edits authentic self-expression. The trained expectation of public numerical approval gradually replaces genuine creative impulse as the primary filter for sharing decisions.
The 2019 Instagram hidden like count experiment produced mixed results. Explicit social comparison anxiety reduced for some users aware of the change; but users adapted by finding proxy metrics (follower counts, comment counts, story views), and the fundamental dynamics of approval-seeking largely persisted. Removing the number does not remove the motivation the number has trained.
The restoration of visible like counts was structurally determined by the business model. Creator backlash, advertiser concerns, and the engagement model's dependence on visible social reward visibility all pushed against the experiment. The compromise — user choice with visible counts as the default — ensures that defaults (which most users keep) maintain the status quo, while technically offering a protective option to those who actively seek it.
The approval economy has a qualitatively different effect on professional creators. For people who earn income from their social media presence, like counts and engagement rates are economic realities, not merely social ones. This converts the psychological dynamics of the approval economy into a labor relationship, where the creator's authentic self-expression is filtered through the economic calculus of what the algorithm will reward.
The creator economy amplifies approval dynamics to an existential level. When social approval equals income, the behavioral patterns described in this chapter — compulsive checking, content optimization toward audience preference, anxiety about underperforming posts — become not merely distressing but economically threatening. The "approval trap" for professional creators is a materially different situation than for recreational users, and it deserves separate ethical analysis.
Platforms didn't invent social approval needs — they digitized, quantified, and weaponized them. The like button did not create the human need for social acceptance. It created a specific delivery mechanism that is quantified (a number), public (visible to all), variable (uncertain reward), persistent (permanent social ledger), scaled (potentially millions of viewers), and algorithmically distributed (rewarding certain content types over others). These specific features are what make it qualitatively different from prior social feedback environments.
Ethical social media design would decouple social connection from quantified approval metrics. The fundamental design question raised by Chapter 10 is whether social connection and sharing can be supported without quantified public approval counts — or whether those counts are so embedded in engagement architecture that removing them threatens the platform's commercial viability. Instagram's experiment suggests the latter, which is itself a significant statement about what the platform was built to do.

Chapter 10 Key Takeaways: Social Rewards and the Approval Economy — Why Likes Feel Like Love