Fig. 6: Correlational similarity between CNN model responses (n = 8) and neural responses in individual human visual areas (n = 8) to clear objects and objects in visual noise. | Nature Communications

Fig. 6: Correlational similarity between CNN model responses (n = 8) and neural responses in individual human visual areas (n = 8) to clear objects and objects in visual noise.

From: Improved modeling of human vision by incorporating robustness to blur in convolutional neural networks

Fig. 6

Leftmost panel shows results pooled across all stimulus conditions; subsequent panels show results for clear objects, objects in Gaussian noise and objects in Fourier phase-scrambled noise; examples of these stimuli are shown above. Error bars indicate ±1 SEM. Two-tailed paired t-tests were performed to determine statistical significance (*p < 0.05, **p < 0.01, and ***p < 0.001, uncorrected; the exact p values and raw values are provided in the Source Data). Elephant image from the authors. LOT, lateral occipitotemporal cortex; VOT, ventral occipitotemporal cortex. LOC, lateral occipital cortex; FFA, fusiform face area; PPA, parahippocampal place area.

Back to article page