Nativism/empiricism debate

Chapter 5 endnote 20, from How Emotions are Made: The Secret Life of the Brain by Lisa Feldman Barrett.
Some context is:

Scientists have debated for hundreds of years over what you’re born with versus what you learn, and I won’t enter that debate. [...] This is the nativism/empiricism debate.

The nativism/empiricism debate is a philosophical debate about whether knowledge is inborn or learned from experience. This animated and important debate covers thousands of years and a wide continuum. On the extreme empiricist end of the continuum, you are born experiencing the world as “one great blooming, buzzing confusion” — perceiving nothing, knowing nothing — and use general learning mechanisms to acquire knowledge about the world. On the extreme nativist end, concepts are built into the structure of your mind.

When it comes to the science of how emotions are made, the nativism/empiricism debate is not central, because there is solid scientific evidence that infants are not born with emotion concepts. Most scholars also admit that different concepts may be formed in different ways, so the origin of one concept does reveal the origin of all concepts. So when it comes to the nature of emotion, any discussion of this vital debate won't change the basic structure of the hypotheses or interpretations of experiments. Still, some points are worthwhile for the interested reader.

A modern nativist hypothesis is that you are born with certain concepts (e.g., for objects, numbers, agency, cause) or with specific, innate learning mechanisms to learn those things within the first few years of life, implying that not all perceptions are constructed. These "given" concepts provide the basis of future learning. A modern empiricist hypothesis, in contrast, is that you are born with powerful learning capacities that are well modeled by the mathematics of Bayesian statistics. These days, most scholars try not to be orthodox in their views. For example, most nativists agree that learning is part of conceptual development in some way or another. What's at issue is which knowledge is learned and how the learning works: are the mechanisms specific to the domain or do they work across all domains?

I am in no way claiming that statistical learning, fueled by words, is the only way for all concepts to develop. For emotion concepts, however, I think it may be the only way they develop. Still, I am not claiming that there are no innate contributions to how children learn concepts. There are clear examples from the animal world where learning from experience is required before animals can perform even the most basic behaviors, despite the fact that they are born equipped with innate endowments (e.g., birds developing the ability to sing). Innate contributions might be things like: an ability to learn that other humans are agents, that humans are sources of information about the world, an affinity for processing speech signals so that concepts can be learned from other people, that speech communicates information about internal mental states like goals, etc. Alternatively, it is also possible that infants come equipped with the capacity to learn how to learn, so that the concepts they learn are determined by the regularities they encounter.

Given the architecture of a nervous system, I do wonder whether valence and arousal might be a form of innate knowledge, although I doubt it has a domain-specific learning mechanism. That being said, I am confident that emotion concepts would not exist without inductive learning. There is essentially no evidence that emotion concepts, themselves, are innate. In addition to being acquired inductively, emotion concepts can be acquired by direct instruction (e.g., psychotherapy).