A quantum-inspired model for human-automation trust in air traffic controllers derived from functional Magnetic Resonance Imaging and correlated with behavioural indicators

From human-human trust to human-autonomous system trust

“…Although Bonferroni correction is widely used in fMRI studies to reduce the chances of Type 1 errors, in the context of this study, where only 4 experimental subjects were involved, use of Bonferroni correction could be far too conservative and limiting. Several important findings could be missed as a result, and that is why p-FDR is preferred for this study (Pushparaj et al, 2021).…”

“…Furthermore, it is also unclear whether this is a confirmation of the hypothesis that trust and distrust are entirely independent constructs. Pilot tests conducted as part of this project yielded results that suggested that trust and distrust can be represented such that they are simultaneously opposite of one another and independent of each other (Pushparaj, Ky, Ayeni, Alam, & Duong, 2021).…”

“…Using the fMRI in any Human-Autonomous Trust study in the context of ATM was entirely unprecedented (Pushparaj et al, 2021). As such, the initial reference point for similar studies had to be obtained from human-human trust studies.…”

“…As such, the initial reference point for similar studies had to be obtained from human-human trust studies. A meta-analysis of the human-human trust studies conducted with the use of the fMRI yielded the following regions of the brain, shown in Figure 5 (Pushparaj et al, 2021), that were strongly associated with trust across different studies with different methodologies (Riedl & Javor, 2012), which advocated for their construct validity. The fMRI has an unparalleled ability to pinpoint the regions of the brain that experience activation for specific cognitive tasks (Smith, 2004).…”

“…The use of objective data to derive the inherent mechanisms of trust became a priority after a comparison between existing measures of trust, which used theoretical constructs of trust that were not always tested for construct validity. The urgency for this approach became clearer after consulting Human-Human Trust literature, where the use of neuroergonomics has led to a better understanding of the fundamental nature of trust (Pushparaj et al, 2021). Current Human-Automation Trust literature has illuminated the factors that influence trust, such as reliability of automation tool.…”