After many years of testing and writing, our eye tracking paper with fingerprint examiners is published in Plos One. It looks at two kinds of mechanisms for missed identification (cursory search and mislocalization) that are hard to address using techniques other than eye tracking. It also features a subject who made 6 false positive errors.
The article is free to access:
https://journals.plos.org/plosone/artic ... ne.0251674
-Tom Busey
Eye tracking research paper published in Plos One
-
tombusey
- Posts: 7
- Joined: Wed Oct 05, 2016 1:28 pm
-
josher89
- Posts: 509
- Joined: Mon Aug 21, 2006 10:32 pm
- Location: NE USA
Re: Eye tracking research paper published in Plos One
I know it's killing the researchers that they can't follow up with the examiner that made six erroneous IDs (including four that were exemplar to exemplar) but:
I'm still reading...
Not certified, not accredited, less than 50% spent on comparisons, makes you want to not have to include this in the number crunching BUT, they do issue reports so you have to.Our data collection procedures are such that we are convinced that these are not clerical errors.
I'm still reading...
"...he wrapped himself in quotations—as a beggar would enfold himself in the purple of emperors." - R. Kipling, 1893
-
Boyd Baumgartner
- Posts: 567
- Joined: Sat Aug 06, 2005 11:03 am
Re: Eye tracking research paper published in Plos One
Not too surprising. The field already cautions against Level 1 exclusions and lack of orientation clues when deciding Exclusions. This corresponds to cursory comparisons and mis-anchoring without anatomical or orientation landmarks like focal points or ridgeflow.
The research seems to be in line with a couple different fields of research:
First behavioral economics like Daniel Khaneman's Thinking Fast and Slow. Namely that the brain seems to operate in two modes. Fast and Slow with the fast mode being characterized by sterotypes/heuristics/etc and the slow mode being characterized by more deliberate processes.
It also lines up with Evolutionary Psychology from the likes of Ian McGilchrist whose basic premises are that the asymmetry of the hemisphere's of the brain correlate roughly to our time being both prey and predator and the two hemispheres represent perceptive differences that correlate to that. The prey aspect being quick and adaptive strategies that helped us survive by being fast, risk averse and safety seeking while the predator aspect is characterized by slow, observant, pattern recognizing and predictive behaviors.
Lastly, it lines up with the Cognitive Neuroscience of Joshua Greene's thesis that moral reasoning is like a Camera. It has two modes; Point and Shoot and Manual. The tradeoff being efficiency vs adaptability. And that framing of decisions can be used to justify just about anything as moral.
I would push back against the idea that accreditation and certification reflect blanket fail safes against error, especially if that error is a function of cognition. Arguably accreditation is a proxy for agencies giving their Examiners a roadmap on how to think when confronted with conflicts in data or between Examiners. However, we've all seen the RSA proficiency results where Accredited agencies and certified Examiners are regularly in the error data. And maybe this just represents the idea that proficiency tests and error rate studies should post SOPs for their tests/studies as an A/B test against ones without SOPs to see if there is a measurable effect.
The research seems to be in line with a couple different fields of research:
First behavioral economics like Daniel Khaneman's Thinking Fast and Slow. Namely that the brain seems to operate in two modes. Fast and Slow with the fast mode being characterized by sterotypes/heuristics/etc and the slow mode being characterized by more deliberate processes.
It also lines up with Evolutionary Psychology from the likes of Ian McGilchrist whose basic premises are that the asymmetry of the hemisphere's of the brain correlate roughly to our time being both prey and predator and the two hemispheres represent perceptive differences that correlate to that. The prey aspect being quick and adaptive strategies that helped us survive by being fast, risk averse and safety seeking while the predator aspect is characterized by slow, observant, pattern recognizing and predictive behaviors.
Lastly, it lines up with the Cognitive Neuroscience of Joshua Greene's thesis that moral reasoning is like a Camera. It has two modes; Point and Shoot and Manual. The tradeoff being efficiency vs adaptability. And that framing of decisions can be used to justify just about anything as moral.
I would push back against the idea that accreditation and certification reflect blanket fail safes against error, especially if that error is a function of cognition. Arguably accreditation is a proxy for agencies giving their Examiners a roadmap on how to think when confronted with conflicts in data or between Examiners. However, we've all seen the RSA proficiency results where Accredited agencies and certified Examiners are regularly in the error data. And maybe this just represents the idea that proficiency tests and error rate studies should post SOPs for their tests/studies as an A/B test against ones without SOPs to see if there is a measurable effect.