The Interactive Effect of Key Audit Matters' Form and Content on Investors' Information Processing and Decision Making: Evidence from an Eye-Tracking Study.

Xin Xu2020 · DOI

TL;DR

It’s not just what auditors say, but how they present it and how much the reader already knows. This study uses eye-tracking and physiological sensors to reveal that "readable" audit disclosures only improve attention when investors are already familiar with the topic; otherwise, an investor's internal professional skepticism is the primary driver of how they process risk.

What Problem Does This Paper Address?

For decades, audit reports were criticized as "boilerplate" documents with little communicative value. To fix this, regulators introduced Key Audit Matters (KAMs)—disclosures intended to highlight the most significant risks in an audit and how they were addressed.

However, prior research on whether KAMs actually help investors has yielded contradictory results. Some studies say they increase transparency; others say they are ignored. This paper tackles the "black box" of information processing: How do the form (readability via graphs vs. text) and content (familiarity with the accounting topic) of these disclosures interact with an investor’s own professional skepticism? The authors aim to understand why more information doesn't always lead to better risk assessment.

Methods

The researchers conducted a 2×2 controlled experiment using a sophisticated "multi-modal" approach:

  • Participants: 78 advanced financial professionals (averaging 21 years of experience), providing much higher external validity than the student samples often used in such studies.
  • Eye-Tracking: A Gazepoint GP3 tracker measured "fixations"—exactly where and for how long participants looked at specific "Areas of Interest" (AOIs) in the audit report.
  • Physiological Monitoring: Participants wore an Empatica E4 wristband to measure electrodermal activity (EDA) and had their pupil diameter tracked. These served as biological markers for cognitive load and "mind wandering."
  • Experimental Design: Participants read two KAMs—one familiar (fixed asset impairment) and one relatively unfamiliar (goodwill impairment). One group saw these in a graphical/table format (high readability), while the other saw them as standard text (low readability).

Datasets

The study did not use a pre-existing large-scale dataset. Instead, it generated a primary experimental dataset from:

  • Experimental Materials: Adapted and simplified from real 2017 audit reports of Chinese listed companies (Beijing Capital Co., Ltd. and SANY).
  • Psychometric Scaling: The Hurtt Professional Skepticism Scale (HPSS) was used to measure the innate skepticism levels of the 78 professional participants.

Key Insights and Ideas

  • The Familiarity Threshold: Readability (using graphs and tables) only increases visual attention when the investor is already familiar with the topic. If the topic is foreign, the "prettiness" of the presentation doesn't matter; the reader is too bogged down by the complexity of the content.
  • Skepticism as a Driver: When investors face unfamiliar topics, their innate professional skepticism takes over as the main driver of attention. High-skepticism individuals will dig into the "Why" of a risk regardless of how it is presented.
  • "Why" vs. "How": The study splits KAMs into the Why Paragraph (why the matter is a risk) and the How Paragraph (how the auditor addressed it).
  • For unfamiliar topics, the "Why" paragraph is the anchor for risk assessment.
  • For familiar topics, the "How" paragraph provides the incremental value that actually changes an investor's mind.
  • Dedicated vs. Absent-Minded Attention: Just because someone is looking at a page doesn't mean they are "processing" it. By using pupil dilation and EDA, the authors proved that "mind wandering" occurs when readers encounter paragraphs that don't match their familiarity level (e.g., a familiar reader skimming the "Why" section).

Key Results

  • Attention Allocation: Readability had a significant main effect on attention ($F=7.627, p=0.006$), but this was heavily moderated by familiarity.
  • Mediation of Risk: For familiar items, attention on the "How" paragraph mediated the relationship between readability and risk assessment. For unfamiliar items, attention on the "Why" paragraph mediated the relationship between skepticism and risk assessment.
  • Biological Evidence: For unfamiliar items, participants' pupil diameters were significantly larger when reading the "Why" paragraph compared to the "How" paragraph (average difference = 0.546, $p=0.001$), indicating they were more "dedicated" to processing the risk description than the audit procedures.

Limitations and Future Work

The authors acknowledge two main limitations:

  1. Numerical Variance: The specific numbers (book values) used in the goodwill vs. fixed asset examples were different. While they controlled for magnitude, different perceptions of specific numbers could influence results.
  2. Material Simplicity: To ensure internal validity, the experimental materials were simplified. Real-world audit reports are far more complex.

Future Work: The authors suggest adding richer, more specific information to both "Why" and "How" paragraphs to see if the effects hold in more complex environments. They also propose further research into capturing "mind wandering" as a standard variable in behavioral accounting research.

Summary updated

Abstract

Using an eye-tracking and electrodermal device, this study investigates how readability and familarity of the Key Audit Matters (KAMs) jointly affect investors' attention and assessment of risk associated with KAMs.We predict and find that only when investors are familiar with the information discussed in the KAM, readability will affect their visual attention to the KAMs.When investors are unfamiliar with KAMs, professional skepticism becomes the main driver of their attention allocation.Furthermore, we distinguish dedicated visual attention and absent-minded visual attention, arguing that the positive correlation only exist between dedicated visual attention on the information and the relative importance of each information in decision-making.Consistent with this argument, we prove that Why (How) Paragraph plays a more important role in risk assessment when investors are unfamiliar (familiar) with the KAM.Our results should be of interest to regulators and auditors as they consider what and how to communicate KAMs in the audit report.
BibTeX
@misc{xu2020interactive,
  author = {Xin Xu},
  title = {{The Interactive Effect of Key Audit Matters' Form and Content on Investors' Information Processing and Decision Making: Evidence from an Eye-Tracking Study.}},
  year = {2020},
  doi = {10.26226/morressier.5f0c7d3058e581e69b05cfd3},
}

Loading related content...

Figure