One Ear Higher Than the Other

The bird is the barn owl, Tyto alba. The finding is R. S. Payne, “Acoustic Location of Prey by Barn Owls (Tyto alba),” Journal of Experimental Biology 54(3), 535–573 (1971), doi:10.1242/jeb.54.3.535. His first summary point, verbatim: “Barn owls (Tyto alba) can locate prey in total darkness using only the sense of hearing, with an error of less than 1° in both the vertical and horizontal planes.”

Both planes is the part that matters, and the whole zine is about the second one. Less than a degree left-or-right is impressive. Less than a degree up-or-down should not be possible with two ears at the same height, and spread six is where that problem gets stated.

What we could not read. Payne’s paper is behind the publisher’s login; we have its summary verbatim from the journal’s own abstract page and nothing beyond it. The two papers this zine actually leans on were read in full, and they are named on the spreads that use them.

This is part one of two and they only work back to back. Part two is on the next spread; do them in one go, same room, same claps.

Keep the claps roughly level with the child’s head and to the left and right, not above and below. Left-and-right is the direction this half is about, and it is the direction people are good at. Up-and-down arrives on spread six, and it is a different story.

This is the same manipulation scientists ran on the owl, which is why the demonstration earns its place instead of being an activity. Knudsen and Konishi’s fourth summary point, verbatim: “Occluding the right ear caused the owl to orient below and to the left of the sound source; occluding the left ear caused it to orient above and to the right of the sound source.”

Read that twice, because the owl’s version has an extra half. Block one of your ears and you lose left and right. Block one of the owl’s and it loses left and right and up and down at the same time — below and to the left, above and to the right. That extra word is the entire subject of this zine, and spread eight is where it is explained.

Say plainly what this demonstration is. It is a demonstration of the principle, not a measurement: your result will vary with the room, the distance and how well the ear is covered, and a child who gets one right by luck has not disproved anything. Do it several times and watch the pattern. The measured version is the owl’s, and it is quoted above.

A hand over the outside of the ear, and nothing in it. Never put an object in a child’s ear for this or for anything else.

The two cues in the read-aloud are the real ones, in the owl and in us. Knudsen and Konishi’s sixth summary point, verbatim: “Both interaural onset time and interaural spectrum are used to identify the azimuth of the sound source. If onset time is not available (as in a continuous sound), the owl can derive the azimuth of the source from interaural spectrum alone, but its spatial resolution is poorer.”

Azimuth is the left-right direction. Interaural means between the ears — the whole subject is differences between two ears, never what one ear can do.

“Sooner” and “different” are the child-sized names for onset time and spectrum. The second one is doing more work than it looks: the head and the outer ear filter a sound differently depending on where it came from, so the two ears receive different mixtures of frequencies, not just different volumes.

The read-aloud is deliberately narrower than the truth, and here is the edge of it. Humans are not helpless above and below — the folds of the outer ear filter sound differently depending on elevation, and the brain reads that. But those are cues from one ear’s own shape, not from comparing two, and they are far weaker than the left-right cues. The zine says the two ears have nothing to compare vertically, which is exactly true, and does not say a person cannot tell up from down at all, which would be false.

This is the problem the owl solves, and it solves it in the way the next spreads describe: not with a better brain, and not with better ears, but by putting its two ears at different heights, so that up and down produces a difference between them after all.

This is not our idea; it is the argument Norberg makes for why owls in particular need this. R. Å. Norberg, “Ocurrence and independent evolution of bilateral ear asymmetry in owls and implications on owl taxonomy,” Philosophical Transactions of the Royal Society B 280, 375–408 (1977) — the spelling of the first word is the journal’s own. Verbatim from his abstract:

“When an owl localizes prey by hearing, the direction of the source usually forms a shallow angle with the ground. Therefore, a certain angle of error usually converts into a longer distance along the ground for a vertical error than for a horizontal error. This is a crucial factor that calls for good vertical localization ability of owls which rely on hearing for localization of food.”

That is a geometry argument, and it is checkable with a torch. Shine one along the floor at a shallow angle and tilt it by a hair: the bright spot travels a long way. Aim it at a wall square-on and the same tilt barely moves it. Worth doing if the drawing does not land.

The anatomy, verbatim from the primary. E. I. Knudsen & M. Konishi, “Mechanisms of sound localization in the barn owl (Tyto alba),” Journal of Comparative Physiology A 133, 13–21 (1979), doi:10.1007/BF00663106 — read at full text: “The ear openings and preaural flaps of the barn owl are asymmetrically positioned within the facial ruff, a structure composed of tightly packed, dense feathers that forms two vertical troughs behind each ear opening… The barn owl’s left ear opening and flap are located high in the left trough, whereas the right ear opening and flap are centered in the right trough.”

Two independent papers agree on which side is which, which is why the drawing dares to commit: Krings, Rosskamp and Wagner, working on the American barn owl Tyto furcata pratincola, write that “the left ear opening in the skin is located higher than its counterpart on the right.”

The drawing is labelled from the owl’s point of view, and that is a real trap rather than pedantry. An owl facing you has its left on your right, so a picture that does not say which it means is telling half its readers the mirror image. Ours says.

In the barn owl this is soft tissue only — skin, flap and feather, with a symmetrical skull underneath. Other owls do it in bone. Spread ten is about that.

Knudsen and Konishi’s fifth summary point, verbatim: “With ruff feathers (facial ruff) removed, the owl continued to localize sounds accurately in azimuth, but failed to localize sounds in elevation.”

This is the strongest thing in the zine and it is worth naming the reason. One manipulation destroyed one ability and left the other untouched. That pattern — a dissociation — is far better evidence than a result where everything gets a bit worse together, because a general decline is what you would expect from an owl that was merely upset. Left-right survived intact.

Why the ruff matters at all: the paper explains that at frequencies of 4 kilohertz and above, the ruff becomes an effective sound reflector, and it is the asymmetrical position of the two ears within that reflector that produces the vertical difference. Below about 4 kilohertz the wavelengths are larger than the ruff and it stops working that way — which is why the owl’s errors were smallest between 4 and 8 kilohertz, and why Payne found the birds depend on frequencies above 5.

The feathers grow back. The paper describes removing ruff feathers; it is a reversible manipulation on a small number of trained laboratory birds, and the zine does not pretend otherwise on the child’s page — it says moved out of the way, which is what was done.

Norberg examined living and freshly dead owls of 16 species. Verbatim from his abstract: “ear asymmetry has evolved independently in at least five lines, represented by the respective genera (1) Tyto, (2) Phodilus, (3) Bubo, Ciccaba, Strix, (4) Rhinoptynx, Asio, Pseudoscops, and (5) Aegolius. Bubo, Ciccaba, and Strix probably represent more than one line of origin of ear asymmetry.” And on what it is for: “bilateral ear asymmetry in owls serves to make the vertical directional sensitivity patterns different between the two ears for high frequencies, thus making possible vertical localization based on binaural comparison of intensity and spectral composition of sound.”

How many times is itself unsettled, and we would rather say so than pick the tidiest number. Norberg’s own paper says at least five. Krings, Rosskamp and Wagner (2018) summarise the very same paper as “at least four times”. Krings, Müller-Limberger and Wagner (2019) say asymmetry “evolved up to seven times independently”. Four, five, seven — the count depends on which owl family tree you use and on how much two asymmetries have to differ before they count as separate inventions. The read-aloud says at least five because that is what the primary says; the disagreement is real and is not hidden.

The independence is not in doubt, only the counting — and the reason nobody doubts it is the sentence the child gets: the anatomy is different every time. Soft tissue in Tyto; skull bone in Aegolius; a difference in the area of the opening in Bubo and Strix. A single inherited invention would look the same everywhere.

M. Krings, L. Rosskamp & H. Wagner, “Development of ear asymmetry in the American barn owl (Tyto furcata pratincola),” Zoology 126 (2018), doi:10.1016/j.zool.2017.11.010. Verbatim: “Micro-CT scans show that in an anatomically defined coordinate system, the ear openings initially appear symmetrically as does the skull as well as the eyes, the nasal openings, the stapes and the squamosum… Soon after their appearance, the ear openings start to move dorso-occipitally. At the developmental stages 36–39, the left ear opening moves faster than the right one. In this way, an ear asymmetry develops within a few developmental stages. The skull and the other anatomical markers remain symmetrical.”

Everything else stays even — skull, eyes, nostrils, and two named bones. Only the ear openings diverge, and only for a few stages. That specificity is what makes it a mechanism rather than a general lopsidedness.

The little owl is Athene noctua, from M. Krings, E. Müller-Limberger & H. Wagner, “EvoDevo in owl ear asymmetry — The little owl (Athene noctua),” Zoology 132 (2019), doi:10.1016/j.zool.2018.10.002. Verbatim: “a small, but significant ear asymmetry occurs in the embryonic development of little owls, despite the presence of symmetrical ears in adults… the asymmetry in the little owl occurs in the same stages at which the asymmetry in the barn owl (Tyto furcata) develops, but in the little owl the asymmetry vanishes shortly before hatching.” It is awake in the daytime and does not need the trick.

The authors’ own reading, and it is a hedge: “We interpret our finding as an indication of a secondarily evolved diurnal activity in little owls. Further, ear asymmetry might be more deeply rooted in the evolution of owls than previously assumed.” Might be is theirs. If that turns out to be right it would change the counting on spread ten, and nobody has settled it.