The psychology of internal audit: Navigating bias, behavior, and decision-making

What is bias, and why it matters to internal audit?

According to dictionary.com, bias distorts thinking, clouds judgment, and leads to flawed decision-making. For internal audit, bias is important because it can impact objectivity, skew risk assessments, and even the effectiveness of audit recommendations.

Bias can occur in different ways during an audit. For instance:

• Information overload bias: When an auditor is working with a large amount of data, it can lead to decision fatigue and reduced focus, causing the auditor to sift through irrelevant information, unconsciously filter out key details, or rely on only the most readily available information.
• Authority bias: This is when an auditor places trust in individuals in positions of power, sacrificing critical thinking and undervaluing insights and data from frontline employees.
• Framing bias: Information delivery (risk or opportunity) can heavily influence how the auditor and stakeholder perceive it.

Between emails, messaging apps, and social media, even the screen you’re currently reading has become a source of attention disruption. According to Maryanne Wolfe, a UCLA Literacy Director, in a recent interview with NPR, we have so much information that we have a built-in defense mechanism to skim, which is the enemy of deep reading. As individuals consume information in short bursts for efficiency, bias can arise, like when stakeholders skim audit reports or auditors rely on summary reports, overlooking key details.

We can start addressing bias by acknowledging that it exists and understanding its impact. If internal audit really wants to be a trusted strategic advisor, then we must be proactive in minimizing bias and make sure our work remains complete, objective, and impactful.

Modern drivers of bias in internal audit

1. The AI paradox: According to a recent Wolters Kluwer survey, AI adoption by internal audit is set to double to 80% in the next 12 months. While AI is beginning to evolve and reshape audit methods and processes, it is still not perfect. AI evolves as it learns, much like an auditor’s progress through experience. As such, AI inherits biases from the data provided (what we teach it), which will increase errors and assumptions that the auditor is relying on. Internal audit must critically analyze AI findings and remain aware of blind spots in the data. Putting overreliance and blind faith into what AI is telling us will result in automation bias, which the auditor uses to produce inaccurate conclusions.
2. Conflicting narratives and stakeholder influence: Internal audit often rely on information provided by stakeholders, but these perspectives can be incomplete or simply their own opinion, without fact. How many of us have spoken to a stakeholder about risks impacting their function, only to determine that they are simply trying to shift blame to others to minimize their roles in control gaps? If you can’t distinguish between fact and opinion, then stakeholders can inadvertently influence the auditors’ judgment.
3. Time pressures and decision fatigue: Tight deadlines and resource constraints are a growing challenge for internal audit teams. During an engagement, the audit team often finds itself compelled to make quick decisions, which leaves them susceptible to reasoning shortcuts. Such behavior is known as heuristic biases, and these shortcuts may place efficiency over accuracy, leading to oversights in complex engagements.

Before moving on, let’s pause for a moment to briefly discuss AI, both as a transformative for internal audit, as well as how it can hinder and help reduce bias in this field.

How AI can reduce bias

• Data-driven decision-making: AI technology uses algorithms to identify trends and anomalies objectively, reducing reliance on human judgment.
• Behavioral insights: AI systems can analyze data to detect behavioral patterns of non-compliance or inefficiencies, giving audit more insight into the “why” behind the data.

How AI can amplify bias

• Algorithmic bias: AI systems are only as objective as the data used to train them. If the data contains historical biases, the AI may reinforce them.
• Over-reliance on technology: An overdependence on AI results, along with insufficient professional skepticism and critical thinking, compromises the foundation of internal audit, as these skills remain essential.

Cognitive bias: Strategies for mitigation

Cognitive bias is a systematic error in thinking that can cloud our judgment based on one’s perception of reality — like assuming only tall people can play basketball. According to the IIA’s Internal Auditor blog, Building a Better Auditor, cognitive bias can influence how auditors interpret data, assess risks, and communicate findings. Here are some common biases:

1. Confirmation bias occurs when audit focuses only on evidence that supports their assumption or hypothesis.
   • How to Address It: Adopt a structured framework in your audit process that requires the auditor to evaluate alternative perspectives and consider outliers.
2. Anchoring bias is when an auditor places over-reliance on the first piece of information they review, such as a prior year’s risk assessment or initial stakeholder input.
   • How to Address It: Incorporate diverse viewpoints across the audit team during planning and reporting to challenge initial assumptions.
3. Availability bias is when the auditor gives importance to readily available data.
   • How to Address It: Utilize data analytics to review all relevant data, not just the most accessible ones.
4. Overconfidence bias involves the auditor overestimating their ability to detect risks or interpret findings.
   • How to Address It: Incorporate peer reviews into the audit process to provide additional scrutiny and ensure that conclusions are well-supported.