【正文】 s level of cognitive processing. Tetlock (1983a) concludes that individuals analyze evidence more carefully and pay particular attention to inconsistent information when faced with predecisional accountability. According to Tetlock and Boettger (1989), predecisional accountability can also reduce judgmental biases. For example, Tetlock(1983b) finds that accountability eliminated primacy effects in a legal decisionmaking task. Similarly,Kennedy (1993) concludes that accountability mitigates recency. Drawing on the accountability literature, FavereMarchesi and Pincus (2020) state, “accountability will motive decision makers to bee more plex, nuanced, and differentiated thinkers as they anticipate objections and engage in preemptive selfcriticism” (2). In a study of dilution effect in fraudrisk assessments, FavereMarchesi and Pincus (2020) find that accountability decreases the occurrence rate of dilution effects. Based on the accountability literature, this study tests the following hypothesis (stated in the alternative form): H1: Asking individuals to justify their impairment assessment will attenuate bias resulting from individual’s prior involvement in the investment remendation. In other words, the hypothesis predicts an interaction between involvement and accountability. METHODOLOGY Participating in the experiment were 89 undergraduate and graduate students from a large southeastern university. The author acknowledges that the use of student subjects may limit the generalizability of the study’s results. However, Brownell (1995) states that for student subjects to be a validation threat,students (1) must differ from the accountants for whom they serve as a surrogates and (2) any difference must interact with other variables to affect the oute. In this study, any differences between the student subjects and accountants who typically make impairment decisions was minimized by ensuring that the students had the same knowledge of the SFAS No. 144 requirements as an accountant may have. In addition, the subjects in this study had an average of years of experience working in accountingrelated positions, which would also help to minimize and Cheng (2020) defend the use of student subjects in an escalation of mitment study by stating: We are unaware of any particular study that has directly pared escalation of mitment between students and managers. However, some studies have reported consistent findings between their pilot study (using students as surrogate managers) and their main study (using managers – ., Booth and Schulz, 1988). (77). The experimental materials consisted of a case revolving around a large entertainment plex including a multiplex cinema, entertainment center, restaurants, and retail stores. The case provided information about the proposed investment in the entertainment plex and its subsequent performance over a 5 year period. The materials also included a summary of the requirements of Statement of Financial Accounting Standards No. 144, including the list of impairment indicators provided in SFAS 144. The case was structured so that subjects would initially have a favorable impression of the entertainment plex project. Section 1 of the experimental instrument outlined the project as it would have looked prior to the decision to invest in the project. The information suggested a positive outlook for the project and all subjects who were asked to make an investment decision remended investing in the project. Subsequent sections of the experimental instrument provided information about the status of the plex after one year, two years, three years, four years, and five years of operation. While a mix of positive and negative evidence was provided in each year’s status report, the thrust of the evidence was to paint a deteriorating picture, particularly in Years 4 and 5. Subjects were instructed to read the case, which presented a proposed investment in the entertainment plex. After pleting their review of the proposed investment, subjects were asked to read the synopsis of the SFAS No. 144 requirements, which included the example indicators of asset impairment. The subjects were told that they could refer back to the requirements at any point in the experiment. In the next section of the experimental materials, the subjects were given a summary of the project at the end of the first year of operations. The summary included mixed evidence concerning the potential impairment of the entertainment plex assets at that point in time. The subjects were instructed to consider the evidence and to estimate the probability that the assets were impaired at the end of Year 1. This procedure was then repeated for Years 2 through 5. Before each decision, the subjects were reminded that they could refer to the requirements of SFAS No. 144 when making their decisions. The experiment was a mixed design that included two betweensubjects factors (involvement and accountability) and one withinsubjects factor (time). The involvement variable consisted of two levels: no involvement in the prior investment decision and involvement in the investment decision. The involvement manipulation was modeled after the involvement manipulation in Brown and Solomon (1987) as well as the personal responsibility manipulations in Jeffrey (1992) and in escalation of mitment studies (., Brody amp。 Achee, 1994。s operations and scan the environment to determine if any indicators of potential asset impairments exist. SFAS No. 144 provides examples of impairment indicators (see Table 1)。 however, the Financial Accounting Standards Board (FASB) did not intend for the list of impairment indicators to be exhaustive. TABLE 1 Example Asset Impairment Indicators Statement of Financial Accounting Standards No. 144 (FASB, 2