WOW, that is by far the most author-packed paper I’ve ever seen outside of astronomy. I guess they’re trying to brute force their way into being seminal? Ngl I’m a fan. Talk about collaboration, there’s a small army behind this document. Seemingly written from a neuroscientific and psychological perspective which is interesting, though there are a few curveballs like the “Kyoto institute for the future of human society”. It’s tragic to see Picard (the foremother of Affective Computing) get only a short shoutout in the same citation as a dozen others, not to mention the almost complete lack of philosophers.

Overall the intro is both fascinating and frustrating. They do crazy stuff like this:

  Such disagreements can only be settled by examining each camp’s take on theoretical virtues (i.e., what qualities make good framework and resulting theory), and comparing each theory’s underlying sets of assumptions (Galilei, 1953; Newton, 1999; Einstein, 1934; Achinstein, 1983; Duhem, 1982; Sober, 2015; Poincare, 2022; Schindler, 2018; Keas, 2018; Ivanova and Farr, 2020).

I guess they want to back up their approach for their empricist friends, but this is just not how philosophy is done, and for good reason: this is more of more like rhetoric than citation. A hip hop shout-out, if you will. If you’re citing Galileo, Newton, Einstein and Poincaré for one and the same sentence, I’d say that sentence either doesn’t need a citation, or the sentence needs to be a paper unto itself. At least a section!

Table 1 is downright insulting. That’s all I’ll say about that.

  To get to this point, 173 researchers from 23 countries came together as a global, interdisciplinary taskforce to examine existing assumptions and approaches in the study of affective constructs. As a preliminary step, a team within this working group performed an exploratory computational linguistic analysis: identifying 3664 words for feelings, sorting them into feeling categories, and characterizing more specific senses for each word (Siddharthan et al., 2018). 
  Guided by the themes that emerged from that initial exploration, twelve teams went on to produce the twelve reviews of this special issue. Each review summarized the state of current behavioral and neuroscientific research with special emphasis on theoretical concerns

1. That’s very impressive, and I’m sure they’ve advanced the field by pulling it off.

2. Choosing to start with a word bank of supposedly discrete phenomenon is a fantastic example of metaphysics (philosophy of science), and why their definition for that word isn’t sufficient.

3. Ah yes, the two human sciences, behavioral psychology and neuroscience… if you wanted to talk theory, maybe loop in the piagetians? They do dominate clinical psychology, after all. There’s probably some Chomskians left. Plus the whole fields of sociology, anthropology, and, again, philosophy…

  What we offer here is not another theory, nor is it a history or review of the field—it is a scaffold of premises that accommodates existing theories by organizing them in terms of a common set of assumptions, and promotes the articulation of new theories. 

“Sir, I’m not guessing the answer, I’m just providing an organizing set of premises!” Of course, this sentence is completely uncited. They do the typical caveats (“this is just a start”, “future work is called for”, etc), but that helps little, IMO. Again, the only word that comes to mind is “insulting”.

  Therefore, the primary purpose of an organism is to ensure its own viability

… I love the autopoetic framework, but here’s another place where they’re out of their depth and just plain wrong. Many, many animals are more meaningfully oriented towards the purpose of propagating their genetic code than preserving themselves as organisms, arguably (e.g. by Nietzsche) including humans. They probably should’ve called a few zoologists — though the psychologists understandably don’t like acknowledging their super-field, just like we computer scientists often don’t like acknowledging discrete mathematics or philosophy of mind.

You can make this assumption, but to do so without even acknowledging counterpoints isn’t as objective or empirical as they present themselves.

  Collectively, these processes move the organism around a new set of states wherein the organism is comfortable—each state as a position on one of many dimensions of anticipated deviation from viability, addressable in many actionable ways. 

I’ve never heard of “allostasis”, and I’m sure it’s a great framework to study with, but this is another fantastic example of an arbitrary choice of focus presented like an empirical finding. I could think of many, many behaviors that would struggle to fit within a mold of avoiding “error states” or seeking a “comfort zone”, other than in the most vague sense of “any decision ends up choosing one thing over another”, which is a mere tautology.

  How they feel has qualities (qualia) (Nagel, 1974; Chalmers, 1997; Silva, 2023)

Another decades-long debate, done and dusted in a single clause! One wonders why they didn’t get Chalmers to review this paper, it’s not like he’s hard to reach… I don’t think he would approve of citing him in this way, at all.

  One major clue that the structure of affective experiences gives us is that they originate from a first-person perspective (self) (Descartes, 1644; Husserl, 2013; Kant, 1908; Searle, 1992; Shoemaker, 2003; Wittgenstein, 1958; James, 1890; -snip of TWELVE other citations-).

Ok I’m done, this is absurd. Sorry for the spam y'all. This is the first paper in a long time that had an emotional effect on me. What irony!

I wish them well, and I’m glad they’re trying to standardize terminology, on some level. I just don’t think they understand what they’re even doing, much less how to go about it in a lasting manner. It at least warms my heart that this is the establishment; seems like a soft target for innovators.