Have you ever noticed a shift in your confidence when identifying species over time? Many of us start with high confidence, only to realize how much we don’t know as we learn more. This is often called the Dunning Kruger effect, whether or not that is correct. It is often associated with a graph similar to this quick one I made:
How has your experience with identifying evolved? This could mean specific taxa, a place you often identify in, etc. Have you found certain groups particularly challenging or rewarding as you learned more? Are there any groups you’ve lost most of your confidence in, or think you’ve mastered? Share your journey and insights!
This curve summarizes it fairly well. Except maybe that the increase after the valley is a bit too steep. I’m still always scared that there is some obscure lookalike I haven’t heard of so that I have to review all my IDs of that taxon I have made so far…
I’m pretty good at recognizing Harmonia axyridis in its various forms, for example, but whenever I ID an observation in a location new to me, I double check the genus or even the entire tribe Coccinellini both on iNat and on GBIF to see if there really are no lookalikes in that area that make a photo-ID impossible (for me). At this point it is probably not necessary (at least in that extent), but I’m too paranoid of making a false ID.
If I was working with specimens, I could see myself in “This is starting to make sense” for several taxa groups I’ve taken an interest. But since we’re mostly limited to photos, I’m often landing in the trough of knowing very little. For every taxa that I learn how to make a confident ID, there’s a taxa that I simply skip because photos don’t get you anywhere.
You’re talking about Sphaeriidae, aren’t you?
Because I will be, but it probably applies to most of the bivalve clades I’ve looked at.
For myself, I’d have the x-axis be time rather than expertise, as I’m not sure my expertise on any taxon is always increasing. Perhaps if I hadn’t kept looking at different taxa to ward off boredom (and the frustration from how hard it is for anyone to photograph something that might be 4mm long and fragile).
I’d also probably a different graph for each taxon, with some (Sphaerium simile) pretty much never losing confidence and others (Euglesa) never getting that high to begin with.
And for some taxa (Musculium), my confidence continues to bob up and down… some days I feel like they all make sense and other times they all look the same to me. And North America has the highest diversity of the (sub)genus with a merefourthreespecies!
Those groups with the fluctuating confidence are the ones where I do a lot of my broad-category IDs… I’ll come back on a day when they’re making sense!
I agree with the curve and the prior response. I tend to use additional tools in iNat to help in the identification process, like recent observances in the area (especially when travelling) to help narrow the field of possibilities. I feel like the iNat suggestions process is getting better (maybe through AI learning?). One of the most valuable tools is learning from the people in my area that seem to be custodians of a region or area (not sure I can name them in this forum), first learning what pictures to take to make a quality observation (e.g. underside of a butterfly is often more useful than a top side), and subtle queues to definitive ID. If there was a way to bottle that ‘tribal knowledge’ and make that available as an enhanced identification guide, that would be amazing.
I would always identify any debris-carrying lacewing larva as Leucochrysa pavida until someone noted that the species only carries lichen fragments. Now I’m a lot more careful, but I can generally still spot a true L. pavida.
I disagree with the right tail of the curve. Once we have it drilled into us that half the taxa we find here “cannot be IDed to species from a photograph,” our confidence level will never recover.
My learning curve would start out at rock bottom with no confidence (but curious) from the left and slowly rise higher to the right with my eastern North American cicada IDs.
Angles matter: Top, bottom, and wings! Especially in the northeast there is a lot of hybridizing going on.
I’ve spent the last few years observing, learning and identifying local moths and bees. I’ve never never hit the “Wow! This is so easy to identify!” point. For moths, it went from I know nothing about this taxon to “WTF these two moths that look nothing like each other are the same species?” Then WTF, these two moths that look the same are different species? Then, how the hell am I supposed to remember how to spell Guenée? I’m now at the point where I can identify the obvious things if I can remember how to spell them.
Yup! Sphaeriidae definitely applies here for me. Whether or not there’s another valley ahead of me, I’m not sure, though. The imaginary curve can vary a lot depending on the taxon and the identifier. For some taxa, after the first trough, I never had enough motivation to learn more. For others, I never really lost confidence (at least, not yet)
Welcome to the forum! Yup, the suggestions (known as the “CV”, short for “Computer Vision”) are being trained roughly each month, adding new taxa and refining the taxa it already has. Maybe it has a learning curve itself!
Welcome to the iNat forum as well!
That’s fair. However, one aspect of most Dunning-Kruger graphs I’ve seen is that, after the valley, the curve doesn’t go higher than the first peak of confidence. This would represent the fact that your confidence level will never really recover
Internalized perfectionism and overconfidence in other fields have dumbed me down. I’m new to the ID’ing journey but I feel like the fear will not subside any time soon.
True. But then, sorting through IDs, correcting your all your previous dandelion IDs from Taraxacum officinale to Taraxacum sp. can leave one traumatised… :P
(Luckily I only had a few when I was told about the difficulty to ID this genus)
Thanks for the info. I’ve seen it associated with the Dunning-Kruger affect (although I’m sure that’s not correct). Now I know what it’s actually called!
In an exam with 45 questions, most people overestimated their performance, and the people who scored the lowest overestimated their performance the most. The highest scorers were the only ones who underestimated themselves, although not by much.
This graph explains a lot of what we see on social media, blogs, etc.
My confidence level does really not change over time, it depends on the resources available. If it’s a bird where every single species has a common name and people have documented every little difference on even the most similar of species, then I can make the assessment really quickly. If it is an insect genus with 40 species, most of which don’t even have Wikipedia page, that’s gonna be a no.
Once I’ve figured out what ID features to trust, I will know what needs to be visible to ID it. Learning the new ID feature and any details of its reliability are the only times my confidence would change.
