Glutamate is the primary "connection" neurotransmitter in the brain ("It is used by every major excitatory function in the vertebrate brain, accounting in total

Glutamate is the primary "connection" neurotransmitter in the brain ("It is used by every major excitatory function in the vertebrate brain, accounting in total for well over 90% of the synaptic connections in the human brain."). DA and SER are far more sparse and unevenly distributed, concentrating in areas with heavy interconnectivity, such as the limbic system and hippocampus. These monoamines are thought to primarily serve in a modulatory role, affecting performance and stability of neural connections in general, in an on-line fashion. By comparison, ANNs have many meta-parameters (parameters besides the actual weights), such as learning rate and momentum, affecting stability and performance. Being able to tune these on the fly would likely improve performance. The "ANNs are not like the brain!" camp like to point out that the ANN is a mathematical model bearing (on one level) little similarity to the human brain, but I think this glosses over the fact that key features (e.g. layers of connections, CNNs) were directly biologically inspired, and we can still glean a lot by studying the best example in our world of a learning system.