Michelle Craske

One of the field’s principal anxiety-disorder researchers, and the interoceptive-exposure page’s presiding figure — but she enters the wiki through a single measurement study, Zoellner & Craske (1999), read first-hand.

She is filed here for the Schandry/Katkin/Ehlers reason: a name doing load-bearing work across the wiki’s clinical pages — interoceptive-exposure rests on her treatment programme, and cognitive-model-of-panic and interoceptive-training-clinical inhabit the literature she helped build — while almost none of that work is in raw/. The one paper the wiki has actually read is the 1999 heartbeat study, where she is senior (second) author to Lori Zoellner. Nothing on this page should be read as a summary of her career, which centres on exposure-based treatment of anxiety and its learning mechanisms far more than on interoceptive measurement.

What the 1999 study establishes

Zoellner & Craske (1999) does three things the wiki did not have first-hand:

• It replicates the founding group difference from outside Ehlers’s lab. Infrequent panickers counted their heartbeats more accurately than controls (44.7% vs 61.7% error at baseline; 35.6% vs 53.0% across trials), and the difference survived covarying self-rated confidence and felt cardiac arousal. Every other version of this finding the wiki holds is Ehlers’s own or a reanalysis of pooled data she contributed to. This one is independent.
• It runs the arousal experiment on the wrong variable. Caffeine raised electrodermal, muscular and metabolic arousal but not heart rate, and accuracy did not move — so the study’s headline conclusion (“accuracy is unrelated to arousal”) holds only for non-cardiac arousal and does not bear on the cardiodynamic confound, which is about the cardiac signal specifically.
• It supplies a within-subject anxiety→accuracy effect that the wiki’s two camps read oppositely. Error fell as state anxiety rose, with no inverted-U. Craske and Zoellner read it as anxiety magnifying attention to the body (the attentional-bias tradition); the validity debate reads the same unsigned-error pattern as the count-inflation artefact. The study cannot decide between them.

The connection the wiki should keep in view

Craske’s larger significance here is as the origin of interoceptive-exposure — deliberate, non-avoidant contact with feared interoceptive sensation, the “most common application of interoceptive evaluation in current clinical practice” per the Khalsa et al. roadmap. That page’s standing gap is that the wiki has read no efficacy trial for it; the treatment’s “established efficacy” phrasing is inherited from Craske et al. (1997) via Farb et al. and softened by the roadmap’s more sober assessment. If a Craske treatment paper ever reaches raw/, it would close one of the clinical side’s most-flagged holes. For now she is, like Ehlers before the 1993 review arrived, a figure the wiki cites more than it reads.