What are your thoughts on setting taxon photos for animals that only show their head?

Over the past few months, I have noticed a growing trend of default taxon photos being set for vertebrates that only show their head. I was just curious to know how others feel about this portrayal, because attempts to set new default photos are reverted back to the photo of their head.

I’m admittedly partial towards photos that show the whole animal in the frame, but I’m curious to know how other users feel.

My thoughts are this:

Taxon photos are used to help users recognize and identify the species they see in the field. In this way they serve a similar purpose to illustrations in a field guide. According to policy, the best default taxon photos have key diagnostic features visible and are visually “attractive” (because they are used throughout the website, mobile apps, and Seek). Additional photos could highlight morphs and other life stages, but the default photo - the first photo - will oftentimes be the only image that users will look at when attributing an ID to their observation. It is important to show as much of the subject as possible within that default photo.

For animals, the best default photo features the animal’s whole body within the frame. For closely related taxa, subtle physical characteristics are the only ways - or at least the best way - to tell certain species apart. This is even more important for related taxa that are sympatric (e.g. ocelot and margay; white-footed mouse and deer mouse; white-tailed deer and mule deer). It also creates a good reference that IDers can cite if correcting an ID, to make sure the user who uploaded the observation would be less likely to make the same mistake again. While photos that only show the animal’s head may look nice, they do not provide the same benefits that a full shot of the animal may.

The great egret serves a good example of why a head profile does not make the best default photo.

This species has a cosmopolitan distribution and is very common. It is sympatric with many other common species within its family that bare a very similar superficial resemblance, including the great blue heron (which has a white morph found in Florida), the cattle egret, and the snowy egret. Observations are often mistakenly attributed to these species. These taxa are best distinguished via the coloration of their bills and the color of their feet and legs. The latter are often considered the best way to distinguish these species because an observer can make out the color of the bird’s legs and feet from a decent distance.

The great egret is also sympatric with more distantly-related cranes and storks, which can be best distinguished by looking at the animals in flight. As a heron, the great egret folds its neck in an “s-shape” when in flight, whereas cranes and storks hold their necks straight.

The headshot currently set as the default of this taxon does not highlight the legs of the animal, so it is not beneficial for users as it could be. A photo that shows the animal’s legs and feet, potentially in flight, would best highlight the necessary features for identifying this species.

EDIT: As pointed out below, the springbok and Thompson’s gazelle are not sympatric, so the example was removed.

8 Likes

I don’t know if/how the photos on the species page can be reshuffled to put a full body photo as the default, but that’s an option. But I suspect that most users would page through the multiple photos of a species to familiarize themselves with the characteristics. Rarely does a single photo show every useful ID feature of an organism.

1 Like

I see your point.
But on another note: full body pictures might often not show the relevant details as well, as they are rather small. I am not sure if I would see the leg detail in your egret example just from the previews.

I personally prefer preview pictures that show me a typical morph (if there is such a thing) of the being, so I can compare with my own pictures. On that basis I narrow it down to possible candidates, which I have a closer look at anyways. So, for me, first picture should give me an impression of “might be it” and further pictures should make the differences more clear.

3 Likes

Thompson’s gazelle and springbok

Just a minor quibble: the two do not occur together, but in geographically disparate areas. I get your point though, that for a lot of people they find it hard to distinguish between the two.

2 Likes

I think that the primary taxon photo for anything should include either the whole organism, or several of the key ID features. If just the head of an animal or one leaf of a plant is enough, fine, but that’s rarely the case. With a few exceptions for organisms that are so widely distributed and easy to identify (think Northern Cardinal) that there isn’t worry about error.

5 Likes

Anyone can alter the gallery of the species (as far as i know)
click on curation ‘curate photos’ and set another default… I would not recommend playing around with it too much but it is an option (i tried, but many of the full body photos are actually quite small…the one currently ‘default’ might work).
To compare photos with mine I would not use the fixed gallery but rather the approved records of the species in the country i’m searching in. It gives more photos -some maybe misID’d-.

1 Like

I often edit taxon photos. If the relevant bit for ID is not in the taxon photos, but there are good research grade photos available, I do add and change the order if it helps.

4 Likes

I can agree with that, but I still feel like a full body view of the animal does a better job of conveying “it might be this” than a photo of just the head.

This photo from Flickr was the original default photo for the great egret, and while it lacks explicit labels like “the feet are always black”, I think it does a really good job of showing the curvature of the neck and an unobstructed view of the feet. I think something like this is more serviceable for IDs than a headshot of the bird.

(I should note that this specific photo cannot be used as a default photo anymore because the photographer changed its copyright status.)

3 Likes

I found a nice flying shot too, but that does not automatically fit the gallery-photo size. So a portrait shot works best I guess.

I have not yet encountered this problem on iNaturalist, probably because most of the time I identify insects, or more specifically moths. Howerver, my opinion is the similar to @astra_the_dragon:

But I also think @Ajott has a good point, too:

For example, going with moths, I curated a while ago the taxon photos for the Peppered Moth and the American cognataria subspecies. In both instances, full body photos are displayed. Also, there are pictures of different morphs/ecotypes, and they are clear enough to tell the distinguishing features for this species/subspecies.

Of course, not all Lepidoptera are easily identifiable from taxon photos, such as species in the genus Phigalia.

I see your point, but also default Great Egret pic is far from the best possible for this species, if we want to distinguish it we need high quality photo, head pattern plays big role in comparing with Intermediate Egret for example, closeups should be at least 2nd pic.

3 Likes

I confess that I never use taxon photos. There are two reasons - I identify Noctuid moths, and the variation between individuals can be extreme. There are a few species where id is matter of fact - Mythimna unipuncta (even this taxon photo is not fully clear on features) comes to mind - but for others (say Euxoa ochrogaster), until one has learned the range of variation, various reliable sources are best. I will often use three websites if I’m not familiar with a species. The second is that I don’t always trust the ID’s associated with a species. Perhaps I’m being somewhat paranoid about that, however.
So in a sense, I don’t have an opinion specifically, but would question the use of a taxon photo as the only source for identification.

2 Likes

Who says only source and for everyone? I think i can help as a first step to get a direction. I personally would also usually research more or leave it a lower level ID if I don´t know too much about a species.

1 Like

This is what I was responding to - perhaps I have interpreted it wrongly.

2 Likes

I’m thinking Vermillion Cardinal. I’m thinking of someone from a state where Northern Cardinals are common, traveling in Venezuela, and seeing what they think is a Northern Cardinal there.

3 Likes

Their heads are very different, so having a close up shot would be helpful, now it’s two generic shots, looks like nobody cared enough to add additional photos for species.

I’m not sure having a hard and fast rule that head shots should be avoided for thumbnails is a good idea. For something like the Cardinal example, I could see a head shot of a male being ok, but maybe not for Great Egret, since most of the ID characters are on the head for the former, but not the latter. I do like having the first photo being a “typical” example of the species, and have run into some issues on that front. For example, the first photo for Eastern Phoebe is of a juvenile, which I have repeatedly changed to an adult, but it keeps getting changed back. Admittedly, the photo of a juvenile is a good photo, but not a good exemplar for the species.

As others have mentioned, it’s definitely worth having multiple photos that show the range of variation of the species, whether that’s different age/sex categories, or different parts of the animal or plant that are relevant to ID. of course, these should only be used as a starting point for ID.

3 Likes

Oh dear, I hope it’s clear that I’m not proposing any actual revision to iNaturalist policy! I think people should have the ability to integrate whatever photos they see to be the best representation for a taxon. Rather, I was curious about how people felt about this type of photo integration in general.

I still personally think that a photo of the whole animal - rather than a specific part of its body - is still the best type of default taxon photo for a vertebrate. I think there is a reason why field guides show entire images or illustrations of animals rather than just their faces. I also think about the default photos set on iNaturalist are the same ones that appear on Seek, and I think it would be a nicer introduction to younger naturalists if the default photo featured the whole animal, but that’s a personal preference, nothing objective.

(I also agree that the best default photo would be of a fully mature individual and of a morph that is most typical, or at lest for vertebral animals. Bonus points if the species is sexually dimorphic and both sexes are within the frame, à la mallard ducks.)

4 Likes

Which just means that iNat can be as agenda-driven as any other wiki. That is one drawback with Wikipedia: you can go in and correct facts, with citations, but if the original writer has an agenda, nothing stops them from going back and undoing everything you did.

I know of several instances where individuals went to taxa and chose a new taxon photo because they took it or a friend of theirs took it and they wanted their photo as the “official” iNat photo.

But the headshot problem is another issue. I see this all the time when I run/judge photo contests, particularly when there is some aspect of public input. A lot of people seem to think that the closer up a photo is for an animal, the “better” the photo is. A full frame headshot is apparently the “best” photo for most people. I’m sure that’s at play here as well.

I wonder if it is the same in plants. Do closeups photos of flowers with no indication of vegetative growth get promoted in place of better diagnostic photos?

Maybe the solution is to have better instructions as to the purpose of the taxon photo. “Remember the photo you choose should show the diagnostic characters of the taxon, not simply be your favorite photo of the taxon.”

3 Likes