{"slug": "one-mechanism-for-many-mental-spaces-a-shared-router-over-a-value-slot-in-models", "title": "One mechanism for many mental spaces: a shared router over a value slot in language models", "summary": "Researchers at MIT found that transformer language models implement a unified mechanism for handling different mental spaces (counterfactual, belief, fictional, temporal) using a shared router and value slot, supporting Fauconnier's mental-space theory over formal semantics' separate-context approach. The mechanism is low-rank, additive with entity identity, and drives inference, with a companion paper exploring belief in depth.", "body_md": "arXiv:2607.10248v1 Announce Type: new\nAbstract: Language builds discourse contexts other than the actual: a painting, a belief, a memory, a hypothetical. Each is a mental space in which the same entity can take a different value, as when a flower is red in reality but purple in a portrait. Formal semantics keeps these contexts apart because their logics differ (modal, temporal, doxastic, depictive); Fauconnier's mental-space theory treats them as one space-building operation. We ask which of these a transformer language model implements, and find a mechanistic version of Fauconnier's unification. The model uses one router/slot format across the inventory: a reusable value slot stores attributed content, and a causally manipulable router (the space index) selects which space is read. A subspace trained with Distributed Alignment Search to control one space type, counterfactual, belief, fictional, or temporal, also controls the others, well above a random floor, on three model families; belief, which formal semantics marks as a distinct case, is not specially separated. The router is low-rank, composes additively with entity identity, and acts through a few late-layer heads. Two further results show the mechanism drives inference and composes: a subspace trained on a rule-derived conclusion flips what the model infers while dissociating from what it reports, and composing space-builders mints a fresh router over the shared slot. This paper establishes the cross-type generality. A companion paper develops belief in depth, because of its special status in philosophy, psychology, and linguistics (epistemology, theory of mind, and propositional attitude reports).", "url": "https://wpnews.pro/news/one-mechanism-for-many-mental-spaces-a-shared-router-over-a-value-slot-in-models", "canonical_source": "https://arxiv.org/abs/2607.10248", "published_at": "2026-07-14 04:00:00+00:00", "updated_at": "2026-07-14 04:34:25.308524+00:00", "lang": "en", "topics": ["large-language-models", "natural-language-processing", "ai-research"], "entities": ["MIT", "Fauconnier"], "alternates": {"html": "https://wpnews.pro/news/one-mechanism-for-many-mental-spaces-a-shared-router-over-a-value-slot-in-models", "markdown": "https://wpnews.pro/news/one-mechanism-for-many-mental-spaces-a-shared-router-over-a-value-slot-in-models.md", "text": "https://wpnews.pro/news/one-mechanism-for-many-mental-spaces-a-shared-router-over-a-value-slot-in-models.txt", "jsonld": "https://wpnews.pro/news/one-mechanism-for-many-mental-spaces-a-shared-router-over-a-value-slot-in-models.jsonld"}}