Humans can follow different social learning strategies, sometimes oriented toward the models’ characteristics (i.e., who-strategies). The goal of the present study was to explore which who-strategy is preferentially followed in the technological context based on the models’ psychological characteristics. We identified three potential who-strategies: Copy the pedagogue (a model with high theory-of-mind skills), copy the engineer (a model with high technical-reasoning skills), and copy the friend (a model with high level of prosocialness). We developed a closed-group micro-society paradigm in which participants had to build the highest possible towers. Participants began with an individual building phase. Then, they were gathered to discuss the best solutions to increase tower height. After this discussion phase, they had to make a new building attempt, followed by another discussion phase, and so forth for a total of six building phases and five discussion rounds. This methodology allowed us to create an attraction score for each participant (the more an individual was copied in a group, the greater the attraction score). We also assessed participants’ theory-of-mind skills, technical-reasoning skills, and prosocialness to predict participants’ attraction scores based on these measures. Results show that we learn from engineers (high technical-reasoning skills) because they are the most successful. Their attraction power is not immediate, but after they have been identified as attractors, their technique is copied irrespective of their pedagogy (theory-of-mind skills) or friendliness (prosocialness). These findings open avenues for the study of the cognitive bases of human technological culture.

人类可以遵循不同的社会学习策略，有时面向模型的特征（即 who-strategies）。本研究的目的是根据模型的心理特征探索在技术背景下优先遵循哪种谁策略。我们确定了三种潜在的 who-strategies：复制教师（具有高思维理论技能的模型）、复制工程师（具有高技术推理能力的模型）和复制朋友（具有高度亲社会性的模型） ）。我们开发了一种封闭式微社会范式，参与者必须建造尽可能高的塔。参与者从个人构建阶段开始。然后，他们聚集在一起讨论增加塔高的最佳解决方案。在这个讨论阶段之后，他们必须进行新的构建尝试，然后是另一个讨论阶段，依此类推，总共六个构建阶段和五轮讨论。这种方法允许我们为每个参与者创建一个吸引力分数（一个人在一个小组中被复制的越多，吸引力分数就越大）。我们还评估了参与者的心理理论技能、技术推理技能和亲社会性，以根据这些措施预测参与者的吸引力分数。结果表明，我们向工程师（高技术推理技能）学习，因为他们是最成功的。他们的吸引力不是立竿见影的，但在他们被确定为吸引者之后，无论他们的教学法（心理理论技能）或友好度（亲社会性）如何，他们的技术都会被复制。