Vision Is for Taking Coordinates, Not for Looking at Oneself: Vision, the Vestibular System and the Posture–Gaze–Attention Triangle (The Kuriyama Protocol, Chapter 3)

Part 8 of a series; the third of fifteen chapter-by-chapter articles. Part 1: 10.5281/zenodo.23587879 · Part 2: 10.5281/zenodo.23587670 · Part 3: 10.5281/zenodo.23526074 · Part 4: 10.5281/zenodo.23609857 · Part 5: 10.5281/zenodo.23610607 · Part 6: 10.5281/zenodo.23682055 · Part 7: 10.5281/zenodo.23684537 If the sole is where force enters, vision and the vestibular system are where coordinates enter. Chapter 3 of The Kuriyama Protocol argues that vision's first job is to establish the world's reference frame, not to inspect one's own body. Orientation is estimated from vestibular, somatosensory and visual channels; precisions add, and the estimate is a precision-weighted mean. Gaze direction sets head orientation and thereby spinal curvature, so gaze is a few-bit setting that replaces joint-by-joint instruction. The correct gaze is task-dependent. Applied to mirrors, the precision-weighted mean drives the visual weight toward 1 and the somatosensory weight toward 0, so body-side errors go unused unless the person can translate the mirror image into somatic terms. Ten sections are covered; two (vergence; mirrors) were registered earlier as Part 1 predictions P1, P4 and P4b and are summarized only. Four gaps found while writing, not stated in the book: - The book's only "verified" number was wrong. Adding 45° external and 45° internal hip rotation to get 90° of pelvic turn is incorrect with both feet planted: both hips rotate by the same angle simultaneously, so one-directional pelvic rotation is limited to about 45°. The verifiable one-directional sum is hip 45° + thoracolumbar 40° + cervical 60° = 145°, plus eyes and below-knee rotation. - "Fixation deprives postural control of information; looking near is bad" conflicts, for quiet standing, with evidence that sway decreases with nearer targets and that the central visual field contributes more (Paulus et al., 1984). The field comparison confounded fixation with head flexion, loss of forward optic flow and fear; the claim is narrowed. - After retracting "90% of neck rotation occurs at C1–C2", the book substituted "C1–C2 initiates rotation" without checking it; it is now a registered hypothesis. - The 1–2% figure for people who can correct movement with a mirror was tagged as measured but derives from the author's oral recollection; relabelled. Five falsifiable predictions are registered (numbered 34–38, continuing the series): - P34. Nodding while fixating an eye-level target brings forward head tilt closer to horizontal, more than the same nodding without fixation. - P35. Changing only gaze height shifts spinal flexion–extension in the same direction, more at higher movement speed. - P36. In a 2 × 2 design (gaze height floor/eye level × target near/far), gaze height drives sway and gait variability; at eye level, near targets give equal or less sway than far ones. - P37. Gaze × task interaction: downward gaze favours flexion tasks, upward or far gaze favours extension tasks. - P38. In the first quarter of a slow head rotation, most rotation occurs at C1–C2, with lower segments joining later. Refutations are welcome. The author has no formal research training. Corrections will be issued as new versions, with earlier versions kept retrievable. The article body is in Japanese. An English summary, the registered predictions and the reference list are included in the document. Contact: contact@medi-gym-revital.com