Hey everyone,
When I upload an observation and get multiple identifications, I often wonder—how do you decide when to trust one over the other? If two people disagree, do you wait for more opinions, or follow one that seems most confident? I’m curious about how others navigate situations where there’s a bit of uncertainty in the IDs.
Thanks for sharing your thoughts!
Use a comment to ask others why they added their ID. They should be able to cite evidence. You can go with that evidnece, or you can just not add an ID if you don’t think it allows you to independently reach a conclusion.
Thanks for the tip! I’ll definitely ask for more info next time and wait until I’m sure before adding an ID.
I often check the ID’ers home page. Most of the time it doesn’t help much but sometimes they’ll describe a background that’s helpful.
If I’m not sure who is right despite the evidence, I usually revise my own ID to a broader taxonomic level or withdraw it all together. Someone who knows more than I can break the tie or discuss further :)
You should do this even if you do not get conflicting IDs. Agreeing should not be used to indicate that you trust the IDer’s expertise or that you appreciate their ID; when you add an ID you are saying that you understand how to ID this species and rule out any similar ones (i.e., not just “it looks right”).
I agree that an identifier should be able to justify their identification based on the evidence presented in that observation. Unfortunately, I see people use fallacious arguments at times, particularly the argument from authority (“I have worked in this area and so know this species” or “I did so-and-so and can be confident in my identification” or worse still “that species does not occur in this area”). If the reviewer is worth their salt they will be able to articulate multiple reasons for an identification beyond just the range of known species. Remember, species do not know where they are supposed to be and given the thermal changes associated with climate change are clearly moving in areas where they once did not occur. So, blindly following species lists is just as bad, in my opinion, as saying ‘it just looks like it’.
This thread looks like it is moving toward academic elitism. I’m going to push back a bit in that someone who is familiar with a given species may reach the point of identifying it by gestalt (“it just looks right”), even if the image does not show the specific features that an “expert” would need. We have discussed this before, for example, in threads like If we took an iNaturalist approach to real life.
It is not elitist to suggest that “it looks right” is not enough if one does not know whether there are other species that look similar. To the unpracticed eye, a suggestion may look like a good fit if one does not know what to look for – and yet it may in fact be something completely different. I have made this mistake myself now and again when trying to evaluate CV suggestions for my own observations of taxa I am completely unfamiliar with.
My experience suggests that it is not uncommon for observers to do this when people suggest an ID. Is it a good starting point? Sure. But comparing images on iNat isn’t much use if only a fraction of the species in an area are represented on iNat, as is the case for many arthropods. Often I find that people compare images and come to wrong conclusions about what features are relevant.
Skilled IDers don’t start out IDing by gestalt; they start out by learning a few species and comparing features and often by working through keys or field guides, or by tagging along with someone else who is skilled and getting tips from them. Being able to correctly distinguish similar species and deduce what differences are relevant just by looking at a few photos without any guidance at all is not a skill that most people have – it is something that needs to be developed, and what works for one taxon is not necessarily directly transferable to another.
It is not elitist to recognize that acquiring specialized skill takes a certain amount of time and effort – whether that skill is playing a sport or instrument, mastering a craft, or learning how to identify a taxon. (This skill does not have to be acquired through formal education or book learning, but I think most of us learn more effectively with some form of guidance.)
Nor is it gatekeeping to suggest that one might be careful about making IDs if one does not yet have the skill to evaluate what one is seeing. There is nothing to stop one from putting in the effort to gain that skill; if others wish to do so, I am happy to explain my IDs or offer tips.
Or when you check the identifier’s home page and it’s there 1st ID.
If it’s an animal of some sort, I know I don’t have the knowledge to ID properly so I withdraw my ID as soon as I notice a differing ID (I guess I wouldn’t if it was clearly wildly wrong - claiming a bird was a fly, for example - but I’ve never had that happen). I don’t add a new ID because I’m not interested in putting in the time to learn enough to ID.
If it’s a plant (my area of interest/knowledge), I query the correction and/or provide reasons why I believe it to be wrong. I have occasionally changed or withdrawn a plant ID because I found myself to be wrong, but never blindly.
Thanks for sharing! I pretty much do the same when it’s outside my area—if I’m not sure, I’ll leave it to others and ask questions. But with plants, I tend to dig a bit more and will ask about a correction if something doesn’t seem right. It’s good to hear how you handle it
I agree the ‘does not occur in this area’ can be fallacious - but your reason of climate change isn’t particularly relevant if it’s a species found on a completely different continent. It’s true that plants can move from America to Australia, for instance - but the probability is low enough that I suspect I’d put the burden of proof on the person trying to prove it was the American species rather than the similar Australian species, not the person suggesting the local species.
And realistically, I think we have to take into account the fact that many (most?) photos on iNat don’t provide the evidence required to follow keys 100%, so if we refuse to take range into account, many more species would never get to species level. I know reaching RG is not the sole aim - but in many cases range seems just as valid a factor to take into account as other characteristics. It’s important to be aware of the possibility of changes in range over time, but things significantly out of range are at least open to question. (Coming from a plant perspective - I can see animals could be different.)
You are correct that continent-level differences are easy to ascribe to ‘not occurring here’. I was mostly thinking of within-continent or country lists of what occurs where as justification for either identifying or excluding ID. I keep thinking back to something someone told me once regarding a particular species that was split into multiple species across the different corners of the Amazon. They said (paraphrasing here) ‘keep in mind that when we described those species and listed their geographic ranges, they were guesses at best and that we really don’t know where the species boundaries are.’
As for keys, I too rarely rely on keys 100% and instead have tried to focus on traits that differ between species and that are likely to be visible in photos on iNat. That is, keys are a nice idea for when the specimen is in hand, but mostly impractical for iNat photo identification and other skills must suffice. So of course species range is a part of that…within reason. My issue is more when range is the ONLY trait used to ID a specimen.
Ouch! Yes, I can definitely see why that might make you suspicious of the value of range as a factor within a country…
And yes, your points are fair. I guess I’ve just seen ‘not in range’ used in more reasonable cases and didn’t like what felt like a complete dismissal of that as an argument.
I think ‘not in range’ is horribly dismissive even though I it at least crosses my mind when identifying. Rather, saying it MUST be this because it’s the only thing in range is missing the point that it would be nice if IDs were made based on evidence in addition to that point. I think we’re talking about the same thing, just different uses of the range argument.
which is a prompt to mentor their second ID, and counting.
Welcome to iNat … and … go
And then the taxon specialists have taxon specialists to refine their IDs.
Exactly. Birds fly, boats float, why couldn’t some Patagonian plant find its way to the Greek island where its lookalikes grow? Geographical arguments are no longer tolerable, they now prove even less scientific than the whole “ID by morphology” lore of yesteryear. It is a pity so few users care to provide (repeated) cytology/genetic info in support of their IDs, while still relying on “keyable” features blind to variability and teratologies - as if we’re still living in the times of Linnaeus and its idealized types.
/s
I get how one could interpret it as dismissive, but from my perspective, “out of range” or “X doesn’t occur in your area” an easy-to-understand explanation, plus it’s correct 999 out of 1000 times. When I correct an observation that I believe is misidentified and out of range, I have found that most of the time the observer never touches the observation again. If I am going to leave an explanation, I’d go for a shorter “out of range” rather than a paragraph on the diagnostic characteristics that will never be read. Of course, if someone asks me for explanation I will gladly give an earful on the diagnostic characteristics. I suspect that most identifiers who comment “not in range” are coming at it from a similar perspective.
Creatures do move! I went to remove a European Robin observation from the Amazon and was surprised to learn that it was known to have traveled there by ship. However, they usually don’t show up where they are not known to live – if enough is known about them, which is often (not always!) the case. Range is indeed enough to dismiss an observation (misidentified or with location mislabeled) unless the photos can be identified with certainty or there’s a clear explanation.
I’m sorry, but if you post a photo of species A that can’t be distinguished from species B, from an area where only species B lives (and the distances are great), I’ll mark it as wrong – as you would mark my observation as wrong in the same situation. (I’ve made these mistakes.)