Any theories as to why so many description section subjects don't match observer's actual ID?

Sometimes I have an idea about what causes people to seemingly choose two different subjects for an observation. If the auto ID symbol is there, I assume the problem is the observer didn’t pay attention to the auto ID. If the observer wants an insect but identifies it as a plant, I assume the problem is the observer couldn’t see the insect while uploading photos. But there are also a lot of observations where the observer states in the description something like “bee on [certain species of plant]” and then adds an ID (not AI) of the plant. So obviously the observer saw both organisms. I’m just wondering if anyone has a theory about why so many observers point out a subject in the description section and then identify a different organism, which causes the observation to be in State of Matter Life a long time while half of identifiers are identifying the bee and the other half of identifiers are identifying the plant. Maybe it would help if the description section were retitled to something like “Description and/or Other Information about the Organism to be Identified.”

Do you have a specific example? I know for Questagame the initial ID comes from somewhere or someone other than the observer.

2 Likes

There are occassions where iNat is used as part of a paper or course, for example on pollinators. Students are asked to make observations of the pollinator and plant, and students being students they don’t always follow the instructions very well, and end up putting observations of the pollinator with mention of the plant in the description.

It could be considered a somewhat “off-label” use of iNat, but it does encourage observation of the natural world…

Ultimately, we have to honour the intentions of the observer, but where those intentions are not clear we can either be free to interpret as we choose, or take the opportunity to start a dialogue. There is always of course the option to just skip past them and leave them for someone else to worry about, which I find myself doing a lot these days :)

4 Likes

I’ve seen some where the observer puts something like “Maple tree” in the description, IDs it as a penstemon, and it’s actually a sunflower. I have no idea.

6 Likes

This is the one I had in mind at the time: https://www.inaturalist.org/observations/38678963 I know I could ask the observer, but as I said, there are many other examples, so I thought it must have something to do with how the system is set up, or what observers see on apps, or . . .

Maybe we users should come up with a consistent way to choose in the Identification Etiquette section, such as always go with the observer’s actual ID or always go with the observer’s indication in the description.

Yes, but my experience is that many don’t answer, or else they just write “Sorry” or something else that isn’t an answer. It makes me think that it’s not clear what should go in the description box. I’m inclined to think identifiers should disregard the description box and just go with the actual current ID of the observer, so that in the example all the identifiers would be focusing on the plant.

Notice I skipped it.

I didn’t see the link at the time I replied :)

looking at the observation, I notice that observer mentions both bee and flower, that it has received IDs for both. The fact that the first ID wasn’t challenged as being for the wrong subject would to me put the subject as the flower. I also note that it is an old photo/observation (2008), and looking at the two observations either side of it (using the prev/next on the observation page) I can see that he uploaded bees or pollinators (how ironic that that was my example!) and in one of those gives the pollinator as subject. I also notice that he does respond to questions, so I’m inclined to view that as him not being too fussed about this issue. I would pick one and put my ID. If someone goes to ID this in 3 months time, and sees a 50:50 split of IDs, perhaps they will mark it “as good as it will get”, because without the observer chiming in with their intentions, it really is just going to be a 50:50 forever!

I’ll just add, that I wouldn’t have changed subject with my ID without a comment specifically asking what subject was, but again, I probably wouldn’t have changed it anyway!

Oh, hang on, observers ID was one of the plant IDs at the start… he has stated his intentions quite clearly in that case!

2 Likes

I only put this example up because someone asked for an example. The facts can be different in different observations, but what the ones I’m asking about all have in common are two different subjects indicated by the observer–one in the description box, and another in their initial or most currect actual ID, so that all identifiers have a claim to be following the observer’s intent.

I would go with their most current ID, as it is easier to make an ID than it is to edit an observations description so I imagine that would be the clarification they would make. It would also be the most recent “statement of intent” as it were.

3 Likes

I guess the bottom line is that due to instructions in classes (as you pointed out) or other personal uses of the description section, the reasons for the ambiguities can’t really be understood clearly or fixed by clarifying the purpose of the description box. So it’s just up to each identifier to decide how involved they want to be in trying to work out the ambiguities. Thanks for the insight. It really hadn’t occurred to me that people might be following instructions to make observations of plants, but to talk about the pollinators in the description box. Now that I think about it, probably insects on plants is what most of these ambiguous subject observations involve.

1 Like

Here’s one of the reasons it almost happened to me:

When I’m not in a good spot for data reception (cell/wifi), there can be a weird thing that happens on the app on the android in which some photos appear to be attached to different descriptions, and may be duplicated, etc. That may be one possible explanation for mismatched IDs and notes.

3 Likes

I’ve definitely seen the ones you refer to though… photo of plant and bee, mentions the plant and bee in the description, IDs as plant, but the observation fields all relate to the bee/pollinator. It can get real confusing!

1 Like

Thanks for that information–I never would have thought of that possibility on my own.

In ZA we have a project “Interactions” for such cases. You duplicate the Obs and ID the one as the flower (Digitalis) and the other as the insect (Apis)
The project then requires you to put the links to the other observation. Check:
https://www.inaturalist.org/projects/interactions-s-afr
and as example:
https://www.inaturalist.org/observations/38289479

1 Like

iNaturalist is just bad at interactions. It’s not designed for it, though it seems to me that very useful data about what organisms are found together could be teased out of many photos.

It personally drives me nuts that there is no really good way to deal with pollination events, which the whole !&#$ planet is interested in. Obviously, the work-around is to post the same image twice, ID both the plant and pollinator, add a linking field and a unique interaction ID to both, and put comments in the description so everyone knows what’s going on.

But that’s ridiculously clunky. People don’t want to do all that, and/or they screw it up and it confuses people like @paloma ;) And worse, if someone posts a pic with a pollination event that they ignore, there’s no way for someone else to ever filter to find it.

Why, oh why can’t we just have a “This is a pollination event” button that automates the duplication process and starts the IDs with “plant” or “insect”, or whatever the AI suggests for a closer ID? It could work kind of like the button for non-wild organisms that anyone can click when someone posts a picture of a lemur from an Ohio zoo. That way, it doesn’t have to all be done by the original poster or not at all.

1 Like

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.