Meta’s TRIBEv2 uses 700 brain scans to engineer viral content<\/figcaption><\/figure>\n<\/div>\nMeta can now simulate how your brain will react to a video before it is watched by a real viewer. TRIBEv2, released by Meta’s Fundamental AI Research (FAIR) team in March 2026, is a foundation model trained on fMRI data from more than 700 volunteers, mapping neural activity at approximately 70,000 points on the cortex. <\/p>\n
According to the company, the end result is a digital copy of the human brain, which helps predict neurological engagement in relation to video, audio and text-based content, and does so with 70 times greater accuracy than its predecessor.<\/p>\n
The model uses three different forms of input, video, audio and text, and matches them to patterns based on fMRI brain scans taken in response to watching videos, listening to podcasts and reading text. fMRI was used to identify actual neurological activity, indicating activation areas within the brain.<\/p>\n
TRIBEv2 applies this information to model how a new piece of content will affect brain areas involved in maintaining attention, arousing emotions, and rewarding activity. If these brain areas are activated by the simulation, the content is more likely to hold the audience’s attention in real life.<\/p>\n
Most notably, the main commercial value of TRIBEv2 is that it allows zero-shot predictions, that is, it can predict brain activity without scanning the same person. This way, content creators and platforms do not need to conduct studies with new participants every time a new piece of content becomes available.<\/p>\n
This model works for a wide variety of people, allowing it to be used outside of a laboratory setting.<\/p>\n
With the results of the analysis provided by TRIBEv2, editors will be able to determine the best possible B-roll, pace content based on cognitive load predictions, and reorganize content in a way that maintains the brain activity signatures most correlated with shares and replays. This is computational neuromarketing on a scale never seen before.<\/p>\n
However, the use of such tools goes beyond mere social media optimization. The ability to predict neural responses to content in video, audio, and written language will have wide applications. <\/p>\n<\/p><\/div>\n