You are more likely to copy others when you are struggling. (fizkes/Shutterstock)
In a nutshell
- We adapt our strategies based on personal success. When people are doing well on their own, they stick to individual learning, but when they struggle, they’re more likely to copy others.
- Success, not social pressure, drives who we follow. People naturally gravitate toward successful peers in environments where copying is helpful, but avoid others when it offers no advantage.
- Flexibility beats fixed strategies. The most successful participants weren’t those who relied only on exploring or copying, but those who quickly switched between the two based on changing conditions.
BERLIN — When our ancestors hunted for food, choosing between exploring new territory or following someone who already found a successful spot could mean the difference between feast and famine. International scientists have now recreated this ancient decision-making process using Minecraft to uncover the reasoning behind when we choose to blaze our own trails versus follow someone else’s footprints. It turns out that adaptability itself, rather than either strategy alone, is the true predictor of success.
This study, published in Nature Communications, feels part video game, part psychological experiment. Researchers discovered we’re constantly toggling between trusting our own instincts and copying others, and we’re surprisingly good at knowing exactly when to switch tactics.
The research team wanted to understand the factors that drive people to either explore independently or learn from others. Their findings revealed that humans are far more adaptable than previous theories had suggested.
For the experiment, researchers invited 128 people to hunt for hidden rewards in a Minecraft world filled with breakable blocks. Sometimes the goodies were randomly scattered throughout the environment. Other times, they were clustered in patterns—find one reward, and others were likely nearby.
“Using a game like Minecraft is useful because it simulates real-life challenges. For instance, since you can only see a small part of the game world at a time, you must choose whether to focus on searching on your own or pay attention to what the other players are doing to learn from them,” says study author Ralf Kurvers from the Max Planck Institute for Human Development, in a statement.
Participants sometimes searched alone, and sometimes in groups where they could see what others were doing. The researchers didn’t just track where people moved; they also monitored exactly what each person could see at any given moment, creating a comprehensive picture of when people chose to watch others versus explore independently.
When We Go Solo vs. When We Follow Others
When people found rewards in the patterned environments, they’d immediately search nearby for more, a strategy that makes perfect sense. But when they struggled to find rewards, they’d increasingly look to others who were successful, essentially thinking, “That person seems to know what they’re doing. Let me see what they’re up to.”
People quickly adapted their strategies. In environments where rewards were clustered, participants who found something good became like magnets, with others gravitating toward them. But in random environments where watching others provided no real advantage, people actively avoided each other, minimizing competition.
Leader-follower dynamics emerged naturally in real time, with successful individuals attracting followers and creating temporary activity hotspots that would dissolve once the rewards were depleted.
Both strategies were driven by the same thing: personal success. When you’re doing well on your own, you stick with what’s working. When you’re struggling, you become more likely to copy others.
The most successful participants weren’t those who relied heavily on either strategy. Success came to those who were most adaptable, quickly adjusting their approach based on changing circumstances.
It’s kind of like when you’re cooking a new recipe, and you start by following instructions exactly (social learning). As you gain confidence, you may begin to improvise (individual learning). If your experiments flop, you might return to carefully following the recipe again.
From workplace dynamics to how information spreads on social media, this could help explain why some groups innovate successfully while others get stuck in unproductive patterns.
Our greatest cognitive asset isn’t our capacity for either individual discovery or social learning, but rather our ability to adaptively switch between them based on how well we’re doing. It’s not which strategy you choose; it’s knowing when to change course.
Paper Summary
Methodology
Researchers created a Minecraft-based experiment where 128 participants hunted for hidden rewards by breaking blocks in a virtual environment. Each person played both alone and in groups of four, and searched in both randomly distributed and clustered (“smooth”) reward environments. What makes this study unique is how they tracked participants—not just recording where people moved, but also what they could see at any moment through an automated visual field transcription method. This allowed researchers to analyze exactly when people chose to watch others versus search independently. Each round lasted 2 minutes, with participants searching through a 20×20 grid of resource blocks.
Results
The study found that people dynamically adjusted both their individual search patterns and social learning based primarily on their personal success rate. In environments with clustered rewards, participants searched more locally after finding a reward and were more likely to be followed by others. In random environments, they searched more broadly after success and actively kept their distance from others. Rather than sticking to fixed strategies, people constantly adjusted based on their performance—those who found rewards became central figures that others watched and followed, while those experiencing failure increasingly observed successful peers. The computational models showed that the best predictor of performance wasn’t relying heavily on either individual or social learning, but rather how adaptively people deployed both strategies in response to their own success and environmental conditions.
Limitations
Although participants adapted their strategies to different environments, they still showed some tendency to copy others even in random environments where it provided no benefit—suggesting potential limits to human adaptability. The researchers note that success-biased copying may be difficult for people to completely unlearn since it’s beneficial in many real-world contexts. The study was conducted using computer interfaces rather than virtual reality headsets, which might have provided more natural movement patterns but created technical challenges. Additionally, the research focused on short-term dynamics rather than long-term cultural learning processes, which limits its application to evolutionary timescales.
Funding and Disclosures
The research was supported by the German Federal Ministry of Education and Research and the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy projects. The authors declared no competing interests. The researchers acknowledged technical support from Philip Jakob, advice from Philipp Schwartenbeck, and feedback from colleagues Ryutaro Uchiyama and Alexandra Witt. Illustrations were created by Erinn Acland.
Publication Information
The paper “Adaptive mechanisms of social and asocial learning in immersive collective foraging” was published in Nature Communications on April 25, 2025 (Volume 16, Article number 3539). The research team was led by Charley M. Wu from the University of Tübingen, with co-authors from various institutions including the Max Planck Institute for Human Development, the University of Marburg, the Health and Medical University in Potsdam, and New York University.
