Alexnet-ABeffect: predicted image features from fMRI data

The 320 files (corresponding to 4 subjects, 8 CNN layers and 10 brain regions) are predicted image features from fMRI data with a trained linear regression model used in the conference paper <b>Categorical Difference and Related Brain Regions of the Attentional Blink Effect by Renzhou Gui and Xiaohong Ji </b>(see reference for full paper). Due to the unexpected damage of subject2-PPA-CNN7.pkl and subject2-PPA-CNN8.pkl, subject2 related image features are all removed.<br>Files are generated using source codes from Horikawa and Kamitani (2017) Generic decoding of seen and imagined objects using hierarchical visual features. Nature Communications 8:15037.