You have no idea what you're talking about.
Actually, he does.
There are 15 billion brain cells in the human brain, all with specific connections to hundreds or thousands of other brain cells.
They have connections, but they aren't "specific" in the sense that every single neuron has "predefined" connections, like, "Neuron#10,386,688,132, you need to connect to the following exact list of other neurons: #9,885,282,110, #12,306,478,264, ... etc."
Instead, large-scale regions of the brain (consisting of hundreds of millions of neurons) grow according to the same "recipe" from the DNA, which results in growth behaviors that, if put into English, would be like, "follow the basic neuron growth pattern, but include 10% more than the standard number serotonin reputake vesicles, while extending an axon in the aft direction to a distance determined by the destination's local hormone gradient, while branching dendrites into surrounding neurons in a preferentially ventral direction until the local concentration of biomarker UYL is depleted."
Gene expression/transcription factors, of course, would determine the various parameters, and for the most part the same basic "growth program" would be used by all neurons everywhere in the brain, while large-scale features would have their growth parameters modified by brain-related analogs of the HOX genes (which determine the positioning of body part placement in most multi-celled animals).
This is why errors in portions of the DNA result in things like brain-wide deficits in a particular aspect of neural growth, instead of, say, a clean gaping hole where a quarter of the cerebellum used to be. DNA controls *global* behavior of cells (even when certain genes are expressed only in certain regions), and does *not* contain "lists" of individualized "instructions" for each one of the vast number of cells in the body.
This is basic embryological development, which is understood pretty well in a broad sense, even though of course there are a vast number of details yet to be determined when it comes to the fine details of every organ and substructure. The above description is based on a ton of real research, and is not speculation.
Covering epigenesis in even moderate detail is way beyond the scope of what can be done in a single post. For a proper treatment, you should really take a course in embryology/developmental biology from a university which offers a good biology degree.
But there are some decent layman's introductions on the internet, for example:
Developmental Biology OnlineTo get a flavor of the kind of research that has been and is being done on this topic, see for example:Rediscovering Biology: Online Textbook: Unit 7 - Genetics of Development
Dynamic Development at a Glance
Gradients That Organize Embryo Development
Epigenetic plasticity and polarity of the embryo BIOL114: Chapter 11. Development: Differentiation and Determination
FLY MORPH-O-GENESIS (amazing)
From radial glia to pyramidal-projection neuron: transcription factor cascades in cerebral cortex developmentThere are databases of known genes and processes involved in the control of embryological development. For example:NONLINEAR MODELING OF EMBRYONIC SALIVARY GLAND DEVELOPMENT and EMBRYONIC SUBMANDIBULAR SALIVARY GLAND DEVELOPMENT (excellent)
Mohawk is a novel homeobox gene expressed in the developing mouse embryo
Role of X-Delta-2 in the early neural development of Xenopus laevis
Sculpting the nervous system: glial control of neuronal development
Essential roles for the FE65 amyloid precursor protein-interacting proteins in brain development
The Interactive Fly: A cyberspace guide to Drosophila development and metazoan evolutionFor example, here is the page from that database listing the 40+ genes found to be involved in neural differentiation: Genes involved in neural differentiation. Here's the entry for one of those genes: Gene name "lola", involved in the growth and guidance of axons. Also see Embryonic origins of a motor system: Motor dendrites form a myotopic map in Drosophila from the same site.
(If they're not connected correctly you can end up with a disfunctional schirzophrenic person.)
If the neurons are not connected "correctly" in the sense of with the correct topology and biochemical properties, yes, but it's not a matter of "if neuron #11,383,987,232 doesn't have a direct connection to #7,362,254,234"...
I'll esimate very low and say that it takes 1 kb of data to store the correct connections for each cell. That's 15 gigabytes of data required just to store information on how brain cells connect to each other, not to mention the design of the brain cells themselves.
Read the above material. It's quite incorrect to assume that each neuron requires its own "blueprint data". It doesn't. Everything that has ever been discovered about embryological development (and that would fill entire libraries) indicates that body (and brain) development is done on a much "higher level" than that -- regions of the body are biochemically tagged with a "map" of organ and tissue positions, and then the cells in each region react en masse by triggering specific differentiation programs for the appropriate regions, based on the biochemical markers (which directly affect gene expression/transcription). There's also a lot of cell-cell interaction and feedback, which again is done on a gross "where am I map" basis, which causes cells in the area to, for example, join up together with their neighbors as tissues, or to extend nerves into surrounding muscle, or to spread a network of blood vessels through tissues which are not yet supplied, etc.
Development is not like an instruction manual where every screw and part has its exact position and assembly operations specified. It's more like the opeation of an ant colony, where each ant has the same "programming", but takes on a different task based on where it is, which chemical cues it runs across from other ants or larva, and which environmental triggers it encounters like food or an obstacle in the underground tunnel and so on. Individual ants don't get or need daily instructions saying "ant 382, your task is to walk north four inches, pick up food morsel, return to hill, descend to level 4Q, hand food to nursery ant, etc." Instead, simple behavioral reflexes (triggered by specific cues) "built" into each and every ant interact with the actions of other ants in the colony and environmental conditions to produce a remarkably flexible and emergently complex cooperative behavior which keeps the colony running smoothly and successfully.
Similarly, the DNA of every cell in the body contains "how to act if you're a cell in the XYZ region" 'recipes', with lists of biochemical responses to be triggered in response to external biochemical cues. The result is the growth and subsequent life processes of a multicellular organism, from a relatively "simple" set of cellular "instructions" (the DNA). Okay, "simple" is relative in this case - it's still very complex, but far simpler than would be needed if you were to try to individually "blueprint" every single cell and give it its own unique "intruction manual".
Even birds have to learn how to sing the species song, build nests, raise offspring.
I grew up alongside a flock of free range bantam chickens and observed them for more than 15 years. None of the hens ever succeeded in raising their first or second batch of hatchlings. They stepped on them while trying to protect them, and killed them all. so much for pure instinct.