The anatomy of a Switch

What the middle of an axis actually contains — context-switchers, genuine mid-points, and how the instrument tells them apart.

Switching is common, not exceptional

Large surveys of BDSM practitioners consistently find substantial fractions identifying outside the pure dominant/submissive poles. Wismeijer & van Assen's Dutch sample of 902 practitioners — still the largest psychological study of its kind — found roughly a quarter of women and a sixth of men identifying as switches, and the identity is ubiquitous in community self-description.

So a typing system that forces everyone to a pole doesn't simplify reality — it misfiles a large minority of it. That's why KDI's axes have a first-class middle band (scores 36–64 read as X) instead of rounding everyone outward.

Two different things live in the middle

A mid-range score can mean two genuinely different psychologies. One is the context-switcher: strongly one pole in one setting, strongly the other elsewhere — their average lands mid-scale, but nothing about them is moderate. The other is the true moderate: mild preferences either way, comfortable in the middle.

A single sitting of a Likert scale cannot tell these apart, which is the exact criticism KDI's expert review raised. The instrument's answer is measurement, not self-report: retake under a different Connection Frame and the per-frame comparison shows exactly which axes moved. A context-switcher's axes travel between sittings; a true moderate's hold still. (Earlier versions offered an 'it depends' answer for this — v0.11.0 retired it as one choice too many; old results that used it still read as context-driven.)

What the donation pool will tell us

Whether the middle band is one population or two is an empirical question, and it's one of the first things the open validation pool can answer: if context-switching is real, mid-band donors who retake across frames should split into a high-movement cluster and a low-movement cluster rather than smearing uniformly. If they don't, the band gets redesigned in a future version — that has already happened once, when v0.11.0 retired the 'it depends' answer in favor of measured per-frame movement.

That is what an instrument improving itself in public looks like: the claim, the test that would falsify it, and the data accumulating in the donation pool.

References

• Wismeijer, A. A. J., & van Assen, M. A. L. M. (2013). Psychological characteristics of BDSM practitioners. Journal of Sexual Medicine, 10(8).