Clarification: Our hypothesis is distinct from the claim that the whole brain—not a single, specialized network—is responsible for language processing. It is important to distinguish LANG from the exportation targets, which can be strongly engaged by non-linguistic inputs.
10/n

The strongest evidence for exportation comes from brain regions that build mental models of other minds—the rTPJ the rest of the Theory of Mind network. These regions are engaged by language that describes mental content.
7/n

If the core language system does not fully process the meaning of a linguistic input, what other systems do? Hypothesis: Regions that build mental models of minds, objects, and places, that store our memories and word knowledge, and that contain our perceptual and motor reps.!
6/n

This rep. is abstract (no longer tied to particular words), but critically, it is an approximation of meaning, derived solely from our knowledge of lang. statistics. Thus, LANG is only capable of *shallow* understanding, and is not the end point of linguistic processing.
5/n

During language comprehension, LANG’s goal is to extract info from word sequences by recognizing familiar words and figuring out how they go together. The resulting representation captures both structural and co-occurrence-based information encoded in a sentence.
4/n

Finally, all cerebellar language regions, but esp. LangCereb3, were similar to LANG in their response profiles and showed strong functional correlations during naturalistic cognition (Expt. 4, n=85).
11/n

We also found that responses in LangCereb3 were modulated by many of the same linguistic properties as LANG (Expt. 3c, n=5). Interestingly, responses in LangCereb3 were not strongly modulated by surprisal.
10/n

…than LANG. This suggests that LangCereb3 processes sentence-level meanings, plausibly inherited from LANG.
9/n

The other three regions exhibited mixed-selective response profiles, responding strongly to language, but also to at least one of the non-linguistic conditions in our battery. These regions may integrate information across diverse neocortical systems.
7/n

One cerebellar language region—LangCereb3, spanning Crus I/II/VIIb—responded selectively to language (mirroring the selectivity of LANG), suggesting that the computations it supports are specifically linguistic in nature.
6/n

We then evaluated the selectivity of these regions for language relative to diverse non-linguistic conditions: motor/articulation tasks, demanding executive tasks, musical stimuli, social/communicative visual stimuli, and semantically meaningful visual stimuli (Expts. 2a-f, n=732).
5/n

Using precision fMRI and a within-participant localization approach, we identified *4* regions of the cerebellum that respond reliably to language across modalities (written and spoken; Expts. 1a-b, n=754).
4/n