Integrated deep visual and semantic attractor neural networks predict fMRI pattern-information along the ventral object processing pathway.

increasingly complex non-linearities of model stage), stopping at the number of knots that gave the best fitting model (based on linear mixed effect model comparisons, as implemented in the lmerTest package).There was a highly significant main effect of ROI ( F = 330.7, p < 0.001) reflecting the fact that overall model fit decreases as one moves from EVC to pVTC to PrC (which could be because of differences in the level of noise in the fMRI data across these ROIs).

also in 132,142 other papers