The experiment reported in this paper tests a theoretical model of experienced and novice auditors' information use. The model, based on social cognition research, posits that when judgements are sequential, the information encountered first affects the processing of subsequent information. Specifically, initial information that is in line with expectations results in more superficial processing of subsequent information than initial information that violates expectations. The judgements of 13 experienced and 26 novice auditors were analyzed to determine whether the model is descriptive of auditors' judgements in an ill‐structured task setting (real estate valuation). Results lend support for the model. The implications of the results and the model's impact on audit effectiveness are discussed.

This study examines novice auditors' knowledge acquisition. I investigate whether low-cost, informal training interventions improve knowledge acquisition in complex auditing tasks. I designed an experiment that tests the effect of two alternative training approaches on audit judgment performance: explanatory feedback (a replication and extension of Bonner and Walker [1994]), and self-explanation of the rationale underlying a judgment of reasonableness. The results suggest that each approach promotes procedural knowledge acquisition, and combining the two approaches provides more benefit than either approach alone. The effectiveness of the self-explanation approach is primarily attributable to the learner's level of reasoning. Adding explanatory feedback to self-explanations has no incremental effect after controlling for the effect of the learner's reasoning. The paper discusses the implications of these results for the design of audit training environments.

We investigate how auditors learn the technical aspects of their professional role while performing client engagements, and how that learning process has been shaped by changes in societal, economic, and regulatory forces. Prior studies explicitly recognize that auditors need social skills and demeanor consistent with professional norms as well as requisite knowledge, but those studies generally focus on the processes through which new auditors are molded toward consistency with social norms. In contrast, we focus on forces affecting the transfer of technical knowledge from supervisor (guide) to subordinate (learner) in the everyday work setting. Our evidence derives from semi‐structured interviews with 30 relatively new and more experienced audit partners at one Big 4 firm, thus spanning multiple “generations” of experience. Results confirm that auditors primarily acquire technical knowledge on the job, through the interactions among individual engagement team members. However, partners express concern about changes in the practice environment that may limit effectiveness of on‐the‐job learning, including characteristics of personnel, the approach to formal training at induction, supervisors' reluctance to provide candid feedback, regulatory and economic pressures, and the increased distraction, and reduced interpersonal contact associated with the use of information technology. At the end of the day, our findings raise implications for practice regarding the difficulty of developing effective learning conditions for auditors in the face of these challenges.

Regulators and researchers express concern about auditors who leave their firms for employment at their clients, due to lingering relationships which might represent a threat to audit quality. These relationships could negatively impact audit quality through undue influences of the client personnel on auditor judgment. We examine how these relationships influence novice auditors during evidence collection. Understanding the effects of alumni affiliation on evidence collection is important because undiscovered issues at this phase may go unaddressed, potentially hurting audit quality. Contrary to most research findings, we find that alumni affiliation can benefit the audit by increasing auditors’ evidence collection. However, we also find, when auditors become depleted, the benefits of alumni affiliation actually reverse, as auditors overrely on the relationship, leading them to prematurely cease evidence collection. These findings have implications for both practitioners and researchers. Data Availability: Data are available from the authors upon request.

Auditors often receive summary information or conclusions from management about account balances or internal controls. They must then gather evidence to assess whether this information is fairly stated. In such situations, management can be considered the “first mover” and the auditor the “second mover.” When auditors are the second mover, they are vulnerable to the curse of knowledge bias—the inability to ignore previously processed information (Fischhoff 1977). Specifically, because information from management could be incorrect or biased, auditors must arrive at an independent evaluation of the item in question (e.g., year-end book values, accounting estimates, or internal controls). This study examines the general issue of auditors being “second movers” by investigating how their awareness of management’s severity classifications of internal control problems influences auditors’ initial assessments of internal control over financial reporting (ICFR) under Auditing Standard No. 2. Our experimental design allows us to determine that management’s “first mover” influence on auditors’ judgments is an unintentional cognitive effect, rather than an intentional use of management’s classifications. We further examine whether cognitively restructuring the ICFR assessment task reduces management’s influence on auditors’ judgments by asking auditors to evaluate and explicitly document the likelihood and magnitude of the effect of an ICFR problem on the financial statements. We find that cognitively restructuring the task mitigates management’s “first mover” influence on auditors’ judgments.