As a first order-of-magnitude estimate: about the same. The reason is that cortical networks are largely recurrent, and hence the number of incoming connections should be similar to the number of outgoing connections.

There are of course different types of pyramidal neurons, some of which also have projections outside of the cortical area they reside in. Two well-known examples are cortico-cortical pyramidal neurons located in L2/3, or pyramidal tract neurons located in L5.

If all you care about are the local connections of cortical pyramidal neurons (and you can hence neglect subcortical outputs etc), one starting point could be the study by Narayanan et al., Cerebral Cortex 2015, investigating local axon projection patterns of excitatory cell types in rat primary somatosensory cortex.

Hope this helps-

best,

Robert

It also depends strongly on the type of neuron. Consider Purkinje Neurons in the cerebellum. Their fan in is in the order of magnitude of 100.000, just gigantic. I was not able to find a precise number, but their output is much less. There are just not enough synapses in the deep cerebellar nuclei (the output

Consider Purkinje Neurons in the cerebellum. Their fan in is in the order of magnitude of 100.000, just gigantic. I was not able to find a precise number, but their output is much less. There are just not enough synapses in the deep cerebellar nuclei (the output regaion) to match all the synapses of the molecular layer of the cerebellum. A back of the envelope calculation gives a fan out of 100 to 1.000 only.

An reverse example might be modulatory neurons of the LC or basal forebrain diffusely projecting to cerebral cortex. Few inputs huge output.