I scanned Anne Fernald's "Human maternal vocalizations to infants as biologically relevant signals: An evolutionary perspective" (in Adapted Mind, pp 391+) about the same time I was treating a depressed mother. One of Mom's cognitions was that her newborn didn't like her but loved Dad.
Fernald describes evidence that the wave forms (rise time, frequency, duration, and volume) of a mother's baby talk is constant in Western cultures. The infant response seems to be independent of language. Neural filtering systems are described in the infant's brain; systems that shunt goo sounds to one sector and reprimands to another. Thus, a compliment that is given in a flat tone may not be detected since vocal quality is the important variable.
I relayed these possibilities to Mom. Mom seemed more hopeful; there was also a challenge to her cognition that her child didn't love her.
The outcome, of course, will be ambiguous. Mom started Prozac 6 weeks ago and already is pepping up. Thus, she may goo more effectively even without my instructions to work on her goo sounds. Hopefully, similar coaching will get Mom to challenge similar ideas she may have that she is an ineffective mother.
One aspect of depression can be a muted affect. Anyone have other depressed mothers with similar communication problems?