Two very successful solutions exist to the problem of vision: biological brains and convolutional neural networks (CNNs). Despite the inspiration of artificial architectures from their biological counterparts, the extent to which the two solutions are comparable remains unclear, with evidence pointing to both core similarities (e.g., hierarchical learning of feature detectors with increasingly complex tuning) and key differences (e.g., very different sensitivity to pixel-level noise). In our work, we present results that bridge these two worlds, showing that prominent information processing trends previously found in CNNs are also present in visual cortex and vice versa. We took inspiration from two recent studies – one showing that, along the layers of CNNs, the intrinsic dimension (ID) of data manifolds undergoes a sharp initial expansion, followed by a monotonic decrease (Ansuini et al; NeurIPS, 2019); and another one showing that, along the rat homologous of the ventral visual pathway, luminance and contrast information are progressively pruned away from cortical representations (Tafazoli et al, eLife, 2017). In our work, we re-analyzed the neuronal recordings of the latter study to measure the ID of object representations across rat ventral visual areas. Concurrently, we measured how the mutual information between stimulus luminosity (or contrast) and units’ activation varied along deep CNNs (AlexNet and VGG-16). We found that the trend of variation of the ID across rat visual cortex displayed the two distinct expansion-contraction phases previously observed in CNNs. In CNNs, we found that luminosity information monotonically decreased across layers, mirroring the increase in ID. Finally, measurements on contrast information revealed how training enhances such information in early CNN layers, while actively discards it afterwards, again in agreement with biological observations. Taken together, these findings suggest a similarly tight relationship between dimensionality expansion/reduction of object representations and reformatting of low-level visual information in CNNs and visual cortex.