The study by Kamitani and Tong investigates whether functional magnetic resonance imaging (fMRI) can decode the detailed contents of a person's mental state, specifically focusing on edge orientation, a fundamental visual feature. Using fMRI, they found that ensemble signals in early visual areas could reliably predict the orientation of stimuli presented to subjects. When subjects had to attend to one of two overlapping gratings, the attention-based bias strongly influenced the ensemble activity towards the attended orientation. This demonstrates that fMRI activity patterns in early visual areas, including primary visual cortex (V1), contain detailed orientation information that can predict subjective perception. The researchers developed a method called "ensemble feature selectivity" to measure feature selectivity from ensemble neuroimaging signals, providing a framework for reading out fine-tuned representations and their subjective contents in the human brain. The study also shows that orientation selectivity is most pronounced in early visual areas V1 and V2, with progressively weaker selectivity in higher visual areas. Additionally, the approach was effective in decoding attended orientation under conditions of perceptual ambiguity, suggesting that top-down attention biases early visual area activity. These findings highlight the potential of fMRI to read out fine-grained representations and subjective mental states.The study by Kamitani and Tong investigates whether functional magnetic resonance imaging (fMRI) can decode the detailed contents of a person's mental state, specifically focusing on edge orientation, a fundamental visual feature. Using fMRI, they found that ensemble signals in early visual areas could reliably predict the orientation of stimuli presented to subjects. When subjects had to attend to one of two overlapping gratings, the attention-based bias strongly influenced the ensemble activity towards the attended orientation. This demonstrates that fMRI activity patterns in early visual areas, including primary visual cortex (V1), contain detailed orientation information that can predict subjective perception. The researchers developed a method called "ensemble feature selectivity" to measure feature selectivity from ensemble neuroimaging signals, providing a framework for reading out fine-tuned representations and their subjective contents in the human brain. The study also shows that orientation selectivity is most pronounced in early visual areas V1 and V2, with progressively weaker selectivity in higher visual areas. Additionally, the approach was effective in decoding attended orientation under conditions of perceptual ambiguity, suggesting that top-down attention biases early visual area activity. These findings highlight the potential of fMRI to read out fine-grained representations and subjective mental states.