Idiosyncratic biases in the perception of medical images

Abstract

IntroductionRadiologists routinely make life-altering decisions. Optimizing these decisions has been an important goal for many years and has prompted a great deal of research on the basic perceptual mechanisms that underlie radiologists’ decisions. Previous studies have found that there are substantial individual differences in radiologists’ diagnostic performance (e.g., sensitivity) due to experience, training, or search strategies. In addition to variations in sensitivity, however, another possibility is that radiologists might have perceptual biases—systematic misperceptions of visual stimuli. Although a great deal of research has investigated radiologist sensitivity, very little has explored the presence of perceptual biases or the individual differences in these.MethodsHere, we test whether radiologists’ have perceptual biases using controlled artificial and Generative Adversarial Networks-generated realistic medical images. In Experiment 1, observers adjusted the appearance of simulated tumors to match the previously shown targets. In Experiment 2, observers were shown with a mix of real and GAN-generated CT lesion images and they rated the realness of each image.ResultsWe show that every tested individual radiologist was characterized by unique and systematic perceptual biases; these perceptual biases cannot be simply explained by attentional differences, and they can be observed in different imaging modalities and task settings, suggesting that idiosyncratic biases in medical image perception may widely exist.DiscussionCharacterizing and understanding these biases could be important for many practical settings such as training, pairing readers, and career selection for radiologists. These results may have consequential implications for many other fields as well, where individual observers are the linchpins for life-altering perceptual decisions.

Super recognizers: Increased sensitivity or reduced biases? Insights from serial dependence

“…However, one could argue that superior processing in SRs may not be (solely) attributable to serial dependence but may rather reflect (a combination of) other perceptual biases. Indeed, various biases continuously shape our perception, including negative adaptation ( Webster, 2011 ; Webster, 2012 ; Webster, 2015 ), stimulus-related biases ( Kosovicheva & Whitney, 2017 ; Wang, Murai, & Whitney, 2020 ; Wang et al., 2022 ), and contextual biases ( Albright & Stoner, 2002 ; Awad et al., 2020 ; Pilz & Lou, 2022 ), as well as biases on an individual basis ( Cappe, Clarke, Mohr, & Herzog, 2014 ; Grzeczkowski, Clarke, Francis, Mast, & Herzog, 2017 ; Shaqiri et al., 2019 ). Extending the present findings, it is possible that SRs may exhibit differences in these effects, which could advance our understanding of mechanisms underlying their perceptual superiority.…”

“…Another possibility is that the performance of SRs may be less impacted by perceptual biases known to affect performance in neurotypical observers. For example, visual perception is characterized by several kinds of misperceptions related to the stimuli ( Kosovicheva & Whitney, 2017 ; Wang et al., 2022 ; Wang, Murai, & Whitney, 2020 ), spatial context ( Awad, Clifford, White, & Mareschal, 2020 ; Bar & Ullman, 1996 ; Manassi & Whitney, 2018 ), and temporal context ( Fischer & Whitney, 2014 ; Kohn, 2007 ; Liberman, Fischer, & Whitney, 2014 ; Webster, 2011 ) to which we are exposed on a regular basis ( Schwartz, Hsu, & Dayan, 2007 ). It is thus conceivable that the enhanced performance of SRs could be attributed to a lack of such biases, which in turn could support the rapid formation of robust, invariant facial representations across highly varied input conditions ( Dunn et al, 2022 ; Nador et al., 2021 ; Nador et al., 2022 ).…”

Serial dependence in visual perception: A meta-analysis and review

“…In addition to history effects that are not serial dependence, there are also many types of stochastic biases in vision that are not serial dependence, such as oblique effects in the orientation perception ( Appelle, 1972 ; Furmanski & Engel, 2000 ; Heeley et al, 1997 ), localization biases ( Kosovicheva & Whitney, 2017 ), idiosyncratic stimulus-related biases ( Wang et al, 2020 , 2022 ), and anchoring effects in numerosity ( Sawyer & Wesensten, 1994 ) and face perception ( Goller, Leder, Cursiter, & Jenkins, 2018 ). These biases can be at the group level or at the individual observer level, but they are not serial dependence and should not be conflated with it because they do not display the diagnostic properties revealed by the meta-analyses here.…”