The extent to which people utilize advice from others differs depending on whether the source of the advice is an algorithm or a human. However, no unifying evidence can be used for advice design. Moreover, the use of advice given at intervals (e.g., 70–90%) has not been fully studied. This study proposed a three-step model of the cognitive process of the use of advice with intervals and conducted a simulation and four behavioral experiments. These experiments showed that differences in advice sources affected the cognitive process in which judges decide whether to update their initial judgment based on the advice; this cognitive process was influenced by the relative weight between their initial judgment and the advice interval. These results suggested that for judges to adjust their judgments, designing advice itself is insufficient and advice must be designed according to the relationship between the advice and judge’s judgments.

Despite computational algorithms outperforming humans in certain tasks, algorithmic advice is less used than human advice (algorithm aversion). Thus, algorithmic advice should be designed to avoid algorithm aversion. However, few studies have discussed the use of advice with an interval (e.g., 60.0 ± 2.0 %), a common format in algorithmic advice. This study confirmed in two behavioral experiments (N = 200) that differences in advice sources lead to differences in advice use, mainly by influencing the step at which the judge decides whether to ignore the advice. Therefore, we proposed to individualize the presentation of advice so that the advice would be such that decreases the rate advice being ignored. Our individualization of the advice presentation focused on the distance between the advice and the initial judgment, a significant factor in advice utilization. Another behavioral experiment (N = 100) confirmed that our proposed advice design overcomes differences among advisors.

Making rational judgments is not always easy. Given that aggregation of the distributed labor force through Internet has become common, a simple and cost-effective solution is needed to improve worker performance. We tested the hypothesis that enforcing a certain decision time boosts job performance by not allowing workers to provide answers within a certain short time after presenting the task. We used the binary judgment tasks, and job performance with various enforced decision times were compared. Two behavioral experiments with physicians (N = 628) demonstrated that job performance was improved by enforcing a one-second decision time; this did not affect the cognitive load of physicians. Furthermore, it was suggested that adding a one-second decision time induced the optimal trade-off between the worker’s performance and cognitive load. Our results show that focusing on resource rationality could lead to simple and cost-effective solutions to real-world problems by boosting workers’ job performance.