Cheekbone prominence: what "high cheekbones" actually measures and how much it matters

"Do I have high cheekbones" sits alongside "do I have a weak chin" and "what is my canthal tilt" as one of the most common self-diagnosis questions in this category, and it is one of the least precisely answered. "High cheekbones" is doing three separate jobs at once: bone height, forward projection, and how hollow the tissue underneath reads. Most content answers all three by eye from a single photo.

Definition

Clinically, cheekbone prominence reduces to the position of the malar point, the point of maximum projection on the zygomatic bone, read against two reference distances: how far out it sits relative to bizygomatic width (the cheekbone-to-cheekbone span) and how high it sits relative to total face height. A "high" cheekbone sits closer to the eye than the mouth on the vertical axis. A "prominent" one projects further forward and outward relative to the rest of the midface. Colloquial language collapses these into one word, but they are two separate measurements.

This is a different parameter from midface ratio, covered elsewhere on this blog. Midface ratio measures overall midface compactness, brow to nose base to chin. Cheekbone prominence isolates one specific point on the zygomatic bone rather than a proportion between three landmarks.

What the research actually found

Mogilski and Welling's 2018 study in Frontiers in Psychology used a conjoint data-driven approach, having 922 raters evaluate composite faces that varied across five sexually dimorphic shape cues at once: jawbone prominence, cheekbone prominence, eyebrow thickness, eye size, and face length. Testing all five together, rather than one at a time, let the study rank how much weight each cue actually carried in a rater's attractiveness judgment. Eyebrow thickness and jawbone prominence came out relatively more salient than cheekbone prominence and eye size.

Little, Jones, and DeBruine's 2011 review in Philosophical Transactions of the Royal Society B situates cheekbone prominence as a genuine sexually dimorphic cue tied to hormone exposure during development, not a myth. But the review's broader read across the literature is that shape cues like this one interact with, and are frequently outweighed by, facial symmetry: several of the studies it covers find symmetry alone explaining more variance in attractiveness ratings than any single dimorphic shape feature in isolation.

The practical read: cheekbone prominence is real and measurable, and the research behind it is legitimate. It is not, on the data, the dominant lever that cheekbone-focused content online implies. It is one input, weighted behind jawline shape and eyebrow thickness, and it matters less on its own than symmetry does.

Bone, fat, and what contouring cannot do

Malar prominence is set skeletally by the zygomatic bone. What changes its visible read day to day is the malar fat pad and buccal fat sitting on top of that bone. Weight change moves this substantially: both fat pads respond to overall body fat percentage, which is a large part of why cheekbone definition is often the first thing to visibly shift during weight loss, with no change to the bone underneath at all.

• Malar or cheek filler: adds soft-tissue volume over the existing bone, projecting the visual point of maximum prominence up and out. Reversible with hyaluronidase.
• Buccal fat removal: the opposite lever, removes fat from the lower cheek to increase the visual contrast between cheekbone and hollow beneath it. Surgical, not reversible.
• Contouring makeup: shifts perceived shadow and highlight along the bone using light rather than tissue. Fully temporary, and has no effect on projection itself.
• Weight loss: reduces buccal and malar fat pad volume together, which is why the same skeletal structure can read very differently at different body fat percentages.

How Facet measures it

Facet's cheeks module locates the point of maximum zygomatic projection from a frontal photo and reports its position against bizygomatic width and total face height, benchmarked to the ranges used in the research above. Because projection has a true depth component, a profile photo sharpens the forward-projection reading the same way it does for chin projection and gonial angle elsewhere in the report.

A full example of how the cheeks module reports this, alongside the other nine modules, is posted at [/sample-report](/sample-report). The citation list behind every threshold, including both studies above, is at [/methodology](/methodology).