Usually, that model is correct and the preceding model in which environmental effects on an individual cause it to adapt and pass on those adaptions to children via the germ cells rather than child rearing is incorrect.
But, we have now learned that epigenetic signals which control the activation of different parts of our genome, can be passed down to future generations via germ cells like sperm without any child rearing transmission of those effects.
This was recently demonstrated in a mouse model where a fear target was generated during the life of a mouse and then transmitted purely via sperm cells to children and grandchildren of the mouse.
Mouse parents learned to associate the scent of orange blossoms with a shock. Their children and their grandchildren startled in response to the scent — a sign of fear — even though they had never smelled it before. Offspring also had more neurons that detect the orange blossom scent than mice whose parents weren’t exposed to the scent. Sperm cells alone can deliver this fear message, study authors Brian Dias and Kerry Ressler of Emory University found. DNA in the sperm cells was imprinted with this fearful association: A gene that codes for the molecule that detects the orange blossom odor carried a chemical stamp that may have changed its behavior.Epigenetics opens up a pathway for hereditary effects not present in the genome itself to be passed on (partially explaining the "missing heredity problem"), can help explain evolutionary adaptations that take place to rapidly to be due to mutation and selection mechanisms alone, and open the door to a whole new class of medicinal treatments, since tweaking someone's epigenome is often far easier than gene therapy that changes genes themselves. Epigenetics gives new life and potential scientific validity to concepts once considered purely pseudo-scientific like "ancestral memory."
For example, experiences of trauma and abuse, particularly in childhood, can leave epigenetic signals and negative symptoms that therapy acting on the epigenome could undo. Similarly, substance abuse can impact the epigenome and epigenetic treatment might permit addicts to undo the physical manifestations of their physical addiction and prevent their descendants from being born with an epigenome preset for rapid addiction to the substances favored by the addicted parent. As these examples illustrate, the focus so far in epigenetics has been on the impact that epigenetics may have in the mental health area.
We don't really know much about what kind of lifetime experiences in an individual leave epigenetic traces, how the epigenome can be manipulated, or what portion of hereditary effects observed in humans and other animals are epigenetic or genetic, respectively. Also, while epigenetic effects can last a couple of generations, they don't appear to persist indefinitely in the absence of reinforcing stimuli in the descendant individuals.
But, the existence of both genetic and epigenetic hereditary factors also further supports the emerging results from twin studies and other studies of heredity that show that child rearing practices of parents have less of an impact on the abilities and personalities of children that conventional wisdom would have us believe - something that can reassure parents anxious about how their own failings as parents might screw up their children, and discouraging for parents who are convinced that exemplary parenting is key to their children's success when really their efforts may have only a marginal impact.
From Science News discussing the paper: B.G. Dias and K.J. Ressler. Parental olfactory experience influences behavior and neural structure in subsequent generations. Nature Neuroscience. Published online December 1, 2013. doi: 10.1038/nn.3594.