关键词: 人类协助, 互联网协助, 过分自信, 思考

Abstract: Without outside assistance, people’s cognitive abilities are highly limited. However, accepting assistance can lead people to overestimate their ability to complete tasks independently. With the spread of the Internet, people have become increasingly dependent on Internet assistance to alleviate cognitive workload. No research has yet explored whether seeking human assistance after “asking for assistance” leads to the same level of overconfidence about future tasks as seeking assistance from the Internet. Additionally, methods suggested in previous studies to reduce overconfidence caused by external assistance tend to increase the time and cognitive cost of solving problems, making them less likely to be recognized and adopted by those assistance seekers. Therefore, we aim to address the following three questions: (1) whether proactively seeking assistance from the Internet leads to higher levels of overconfidence about future tasks than proactively seeking help from others; (2) whether a lack of reflection can lead to higher levels of overconfidence; and (3) whether a “thinking tip” can reduce the seekers’ overconfidence in their ability to complete tasks independently in the future after receiving assistance.
Study 1 used a single-factor between-subjects design to explore whether receiving human assistance leads to lower overconfidence than receiving Internet assistance after asking for assistance. One hundred and twenty-six college students were recruited online and randomly divided into three groups. In the first stage, participants were required to answer 10 general knowledge questions as correctly as possible. The Internet assistance group could use “Baidu search” for assistance, the human assistance group could ask the experimenter for answers, and the no-assistance group had to answer independently. In the second stage, participants estimated how many questions of similar difficulty to those in the first stage they could answer correctly on their own. In the third stage, participants in each group independently completed an additional 10 questions of similar difficulty to those in the first stage. Study 2 used a 2 (Internet assistance vs. human assistance) × 2 (thinking vs. no thinking) between-subjects design to explore the reasons for the difference in how various types of assistance lead to overconfidence in future tasks. It also tested whether a more feasible “thinking tip” could reduce the overconfidence caused by receiving assistance. One hundred and ninety-two college students were recruited online and randomly assigned to four groups. Participants in the no thinking groups sought help immediately after reading the questions, while participants in the thinking groups were asked to continue thinking about the question when they sought help. The other stages mirrored those in Study 1.
Study 1 showed that overconfidence in the Internet assistance group was significantly higher than in the human assistance and no-assistance groups. Study 2 showed that “thinking tip” could effectively reduce overconfidence in those who received Internet assistance, but had no effect on the overconfidence in those who received human assistance. In summary, (1) after seeking assistance from the Internet, people show higher confidence in their ability to complete tasks without assistance in the future, and are more overconfident than those who sought assistance from human. (2) The lack of thinking about problems during the process of “asking for help” by individuals who seek assistance from the Internet is an important reason why they have higher overconfidence after receiving Internet assistance than after receiving human assistance. (3) Providing individuals with the tip to think about problems before receiving assistance can reduce the overconfidence in those who receive Internet assistance in completing similar future tasks without assistance.

Key words: human assistance, Internet assistance, overconfidence, thinking