Jennifer Murphy

A Royal Holloway psychologist in the Bird/Catmur interoception lineage — the group behind the claim, load-bearing on alexithymia, that alexithymia is fundamentally “a general deficit of interoception” (Brewer, Cook & Bird 2016). The wiki had carried her by citation only: Murphy, Catmur & Bird (2018), “Alexithymia is associated with a multidomain, multidimensional failure of interoception,” is the source of the multidomain-failure phrasing that appears wherever the wiki discusses whether interoceptive deficits generalize. Her 2022 commentary is the first paper of hers in raw/, so the page is created under the load-bearing-figure convention (as with Schandry, Katkin, Ehlers).

The through-line: measure interoception properly, then ask what it is for

Murphy’s programme has two moves the wiki should keep distinct.

The measurement move (2018). Interoception is multidomain and multidimensional, and studies that measure one cardiac task and conclude about “interoception” are making an unlicensed generalization. This is the same worry Banellis et al. (2026) later confirmed psychophysically and Desmedt et al. made a formal argument — Murphy’s group stated it early, from the alexithymia side.

The use move (2022). Having insisted the field measure ability carefully, she then argues ability is not the whole story: propensity to use interoceptive signals is a separate individual difference the taxonomies omit. This is her distinctive contribution to the wiki — the internal-vs-external axis as a person-level disposition, and a generalization of Dunn’s gain-term result from accuracy to deployment. See is-more-interoceptive-awareness-better.

Placement

Murphy responds directly to Garfinkel (the 2022 commentary is on Suksasilp & Garfinkel’s model), sits beside Desmedt on the measurement-critic axis, and beside Allen in holding that interoceptive ability may not generalize across channels. Thin page — one study in raw/ — and the primary Bird/Catmur sources (Murphy, Catmur & Bird 2018; Brewer, Cook & Bird 2016) remain unread here, so the multidomain-failure and general-deficit claims are held at one remove.