Detecting Financial Misreporting: An Application of Transdermal Optical Imaging
Project: Research
Researcher(s)
Description
Financial misreporting is a first-order concern for a wide range of stakeholders. For example, Enron’s shareholders lost billions of dollars, eroding investor confidence in the capital markets and undermining the SEC’s objectives. Therefore, it comes as no surprise that researchers have devoted tremendous effort to develop and improve models to predict financial misreporting. For decades, these models have relied mainly on accounting numbers collected from the financial statements (e.g., Jones 1991; Dechow, Sloan, and Sweeney 1995; Dechow and Dichev 2002; McNichols 2002; Kothari, Leone, and Wasley 2005; Dechow, Ge, Larson, and Sloan 2011). Recently, a growing strand of research takes a very different approach and builds on findings from the social psychology literature which suggest that liars exhibit many different “markers” that can be used to identify deception such as word usage (Larcker and Zakolyukina 2012) and vocal cues (Hobson, Mayew, and Venkatachalam 2012). Prior research has acknowledged that physiological changes (e.g., blood pressure and heart rate) are particularly effective at detecting deception, but data limitations have precluded this possibility. However, recent innovations in medicine and computer science make this a viable option. In this proposal, we explore the efficacy of a new technology called Transdermal Optical Imaging (TOI) in detecting financial misreporting. TOI was invented by Kang Lee (Tier 1 Canada Research Chair at the University of Toronto). TOI relies on the following intuition. Human skin is translucent, so changes in facial blood flow lead to changes in skin pigmentation. For example, individuals blush when they are embarrassed. The human eye can detect extreme changes in skin pigmentation, but computers are needed to detect subtle changes. Due to improvements in the resolution of cameras, TOI can be applied remotely using modern video recordings. For further details on TOI and its potential in detecting deception, please refer to Kang Lee’s Ted Talk: https://youtu.be/6diqpGKOvic?t=332.Detail(s)
Project number | 9043602 |
---|---|
Grant type | GRF |
Status | Active |
Effective start/end date | 1/01/24 → … |