Lab Rats With Smartphones: Meta's History of Experimenting on Users

In 2014, researchers revealed that Meta had conducted a massive psychological experiment on nearly 700,000 Facebook users, manipulating their News Feeds to show either more positive or more negative content and measuring the effect on their subsequent posts. The study, published in the Proceedings of the National Academy of Sciences, found that emotional states were contagious through social media — users exposed to more negative content posted more negative content themselves. The experiment was conducted without informed consent, without ethics board approval, and without users' knowledge.

The Emotional Contagion Study

The study's methodology was straightforward: for one week, Facebook's algorithm was modified for selected users to filter either positive or negative emotional content from their News Feeds. The result was a measurable shift in subjects' emotional expression — people shown more negative content became measurably more negative in their own posts. The implications were disturbing on multiple levels: Meta had demonstrated both the willingness and capability to influence the emotional states of millions of people simultaneously, and had done so without any consent mechanism or ethical oversight.

The Broader Pattern

The emotional contagion study was not an isolated incident but the most visible example of Meta's ongoing experimentation on its user base. Former employees have described a culture of continuous A/B testing where every aspect of the user experience — from notification timing to content ranking to UI design — is tested on randomly selected user segments. While A/B testing is standard practice in technology, Meta's experiments frequently cross the line from product optimization into behavioral manipulation, testing how changes in content presentation affect user emotions, beliefs, and behaviors without disclosure or consent.

The regulatory response was muted. While the study prompted outrage from privacy advocates and ethicists, no regulatory action resulted. Meta's terms of service were retroactively interpreted to cover research use, and the company updated its data use policy to explicitly mention research — a change that legitimized future experiments rather than preventing them. The absence of meaningful consequences has allowed Meta to continue treating its user base as an ongoing research population, testing behavioral interventions at a scale that no academic research institution would be permitted to conduct.

Users should understand that every interaction with Meta's platforms potentially contributes to experiments they have not consented to and may never be informed about. The ethical frameworks that govern research on human subjects in academia and medicine do not apply to technology companies, and efforts to extend these protections to digital platforms have made minimal legislative progress.