I am super excited to share the recent bioRxiv preprint of the second part of my PhD work at @BoninLab @__NERF!

How is such specific wiring of long-range pathways established during neurodevelopment? The next big question to answer. 🧐 #Neuroscience

These specific and diverse long-range inputs provide essential ingredients for specialized visual representations and functions of HVAs.

The superficial layers of the cortex and the thalamic neurons follow ‘target-specific’ mode, where neurons send information aligned to the tuning of the target HVAs; whereas the deep layers of HVAs ‘broadcast’ information across cortical areas.

Like us, the mouse sees through a series of processing stages with the primary and higher-order visual cortical areas (HVAs) specialized for the most complex visual features. Using in vivo circuit mapping and functional imaging, we uncovered two modes of interareal communication.

‘Higher-order cortical and thalamic pathways shape visual processing streams in the mouse cortex’.

Have you ever wondered how your brain processes what you see? Our work in press at Current Biology uncovered rules governing complex wiring of the visual cortex that underpins visual functions. @BoninLab @__NERF

Big thanks to all the support from @BoninLab @__NERF @LBI!

In sum, we found areal and cell-type specific wiring rules of intracortical and thalamocortical pathways in mouse higher-order visual circuits. We believe similar rules might be universal across mammalian species to construct the common framework of visual cortical network. 10/10

While HVAs integrate specific cortical inputs but these inputs seem insufficient to fully explain the high specificity of HVAs. We looked at another major information pathway, the projections from the thalamic LP, and found highest target-specific in all pathways invested. 9/10

Is target-specificity a universal feature of intracortical projections? Unlike L2/3 neurons, L5 HVA neurons broadcast information biased to the source areas. Hence there are multiple parallel communication pathways mediated by different layers and cell types. 8/10

Looking at the output of individual HVAs, L2/3 projection neurons in HVAs are functionally-target specific. This target-specificity was similarly observed in L2/3 projection neurons from lower to higher visual areas, suggesting a common rule of L2/3 intracortical pathways. 7/10

Do individual HVAs receive diverse or similar information from other cortical areas? We found AL and PM, which comprise heterogeneous functional populations, receive diverse inputs. Whereas area A, functionally homogeneous, receives homogeneous inputs. 6/10

HVAs receive dense anatomical inputs from functionally-like areas and avoid input from functionally segregated areas. 5/10

We first characterized the diversity and specificity of visual neurons (70000+ cells) within and across V1 and HVAs. We saw profound functional diversification between areas, and local heterogeneity within areas. 4/10

We characterized all major input pathways to three functionally distinct HVAs (AL, PM and A) in the mouse, including the inputs from lower and higher-order visual cortex and from the higher-order visual thalamus LP. 3/10

Here we asked in the mouse visual cortex (for some reason we love this model): to which degree can specialized visual representations in higher-order visual cortical areas (HVAs) be explained by the long-range inputs they receive? 2/10

Functional specialization of cortical areas is a fundamental feature of the mammalian cortex. This is believed to arise from converging inputs from other cortical and subcortical sources. However the contribution and importance of different input pathways are not clear. 1/10

Following up on this study, we ask: How are these specialized visual areas and streams involved in visual perception and behavior (foraging, navigation, learning …)? How are long-range communication pathways organized to support specialized visual representations and functions?

Full datasets and analysis code are available here: