Skip to content

“Worthwhile content in PNAS”

Ben Bolker sends an email with the above subject line, a link to this article, and the following content:

Experimental evidence that hummingbirds can see purple … researchers used Stan to analyze the data …

The article in question is called “Wild hummingbirds discriminate nonspectral colors” and is by Mary Caswell Stoddard, Harold Eyster, Benedict Hogan, Dylan Morris, Edward Soucy, and David Inouye.

All four of the figures in this paper are pretty. I’ve just shared two of them.

Here’s the key bit:

We conducted a series of field experiments on nonspectral color discrimination in a population of wild, free-flying broad-tailed hummingbirds . . . we trained hummingbirds to associate one color (the reward color) with a sucrose solution and a second color (the nonreward color) with water . . . Each platform supported a tube and a clear plastic saucer partially filled with either sucrose solution or water. . . . Because hummingbirds can readily learn location cues, after each trial we swapped the locations of the reward (sucrose saucer with reward color) and nonreward (water saucer with nonreward color) setups to prevent birds from cueing on the spatial location of the reward.

It’s the bottomless soup bowl for hummingbirds! Maybe this experiment will be featured in the next edition of Nudge.


  1. Anoneuoid says:

    Off topic but heres a good one for the blog:

    Lay coverage here:

    “AKL-T01 is an investigational digital therapeutic that uses a proprietary algorithm designed to improve attention and related cognitive control processes, by training inter ference management at an adaptive and personalised high degree of difficulty. Interference is instantiated through a video game-like interface displaying two tasks that are to be done in parallel (multitasking):

    *a perceptual discrimination targeting task in which users respond to the instructed stimulus targets and ignore the stimulus distractors (similar to a Go–No-Go task)*

    and a sensory motor navigation task in which users continuously adjust their location to interact with or avoid positional targets


    In the current study, there were no differences between AKL-T01 and the control condition on secondary measures, and several factors might explain these f indings. First, it is possible that parent or clinicianreported outcomes (ie, ADHD-RS) are not sensitive to the effects of AKL-T01. In other words, the shown effects of the intervention on attentional processes may not be as readily observable by parents and clinicians.”

    This is the only test they improved on:

    So children got better at go-nogo tasks after practicing for a month, but otherwise seemed the same. No mention of this very likely explanation is made. FDA approved it for ADHD.

  2. Jonathan says:

    1. Bottomless soup bowl appeared in a boys book. I cant remember the series; it was a super clever person who would take on challenges, some far-fetched. In one, he was invited to dinner and the spaghetti would keep filling. He wasnt fooled: he showed, if I remember, how the table was a converted pool table and that legs had pumps. Silly.

    2. Turtles like purple. My dad would bring home old science movies and we’d watch them on the bedroom wall with a little projector. One showed that turtles preferred bread of certain color. Cue the turtles. See them walk. See them pick the purple. Cut to shot of turtle with purple bread in its mouth. No idea if they even did statistical analysis.

  3. jd says:

    They did their experiments in the summer in Gothic, Colorado. Hey I think I’ve been there! I think that is just up a dirt road out of Crested Butte. Gorgeous place to work (in summer)…I’m definitely in the wrong line of work. I should be designing hummingbird experiments.

  4. David J. Littleboy says:

    “Nonspectral” colors: is there any such word?

    It seems that it’s old news that birds (and some other animals) have UV sensitive cones. Which means that they can’t not see (i.e by definition can see) “colors people can’t”. So why isn’t this article titled “Something we already knew turns out to be true”???

    (Irrelevant aside: Color is hairy because since we have only three color receptors, we can only represent a subset of the possible colors that might be possible were spectra arbitrary. But since the spectra of ambient light tends to be generated by black-body radiation (not large numbers of monochromatic LEDs at narrow intervals at random levels), there’s no problem (i.e. nature usually doesn’t play tricks on us). Heck, there have been color imaging systems that use only two colors and aren’t all that bad. (It’s more complicated than that, since overall intensity gives a forth variable that’s somewhat related to color saturation. And to add insult to injury, Japanese has words for (two of three possible) muddiness/color purity concepts that only appear in very technical articles (濁色 (a hue to which some amount of both black and white have been added), muddiness), 清色 (a hue to which some amount of white or black has been added (note that this covers two ideas), vividness in non-technical language, but that’s a different thing)).) But to be somewhat relevant, everything in color science assumes three cone/primary colors, so it isn’t going to work for beasts with four cone/primary colors.)

    There was a similarly obnoxious article the other day to the effect that human peripheral vision color is actually inferred color, presented as some sort of amazing revelation. But we already knew that there are very few cone cells outside the center of the retina, so the gestalt of the colorful world we think we are perceiving has to be a constructed gestalt.

    I wonder if Andrew has a word for this, i.e. something presented as an amazing revelation that logically has to be true or that we already knew?

    • Adam Wheeler says:

      According to the article, nonspectral colors are those which can’t be induced with monochromatic light, or, more accurately, one in which non-adjacent cone types are stimulated. I agree it’s a confusing name.

      The significance of the article is that they have verified that hummingbirds actually perceive color in a 4D space, and use this information in a non-laboratory setting.

      I’m not a biologist, so I’m really not in a position to judge the “duh-factor” of this paper, but the authors do delineate carefully what is new here.

      • David J. Littleboy says:

        “The significance of the article is that they have verified that hummingbirds actually perceive color in a 4D space, and use this information in a non-laboratory setting.”

        Ah, that’s actually a real/useful/interesting result. Thanks for taking the time to figure it out!

    • Martha (Smith) says:

      David said, “I wonder if Andrew has a word for this, i.e. something presented as an amazing revelation that logically has to be true or that we already knew?”

      Perhaps, “cogito, ergo sum”?

Leave a Reply