A new AI model mimics human thinking with striking accuracy, even in unfamiliar scenarios.
Researchers at Helmholtz Munich have created an advanced artificial intelligence system capable of mimicking human decision-making with impressive precision. The model, named Centaur, was trained using data from more than ten million decisions collected through psychological studies, allowing it to generate responses that mirror human behavior in realistic ways. This breakthrough offers new possibilities for deepening our understanding of how people think and refining existing psychological frameworks.
For years, the field of psychology has sought to fully capture the intricacies of human thought. However, past models have typically been limited to either explaining how people think or predicting how they act, rarely managing to accomplish both.
Led by Dr. Marcel Binz and Dr. Eric Schulz from the Institute for Human-Centered AI at Helmholtz Munich, the research team has now introduced a model that bridges this gap. Centaur was trained on a comprehensive dataset known as Psych-101, which compiles over ten million decisions from 160 different behavioral experiments.
Centaur stands out for its ability to anticipate human responses not only in familiar contexts but also in brand-new situations. It recognizes recurring decision-making patterns, adjusts to new environments with ease, and can even estimate reaction times with a surprising level of detail.
“We’ve created a tool that allows us to predict human behavior in any situation described in natural language – like a virtual laboratory,” says Marcel Binz, who is also the study’s lead author. Potential applications range from analyzing classic psychological experiments to simulating individual decision-making processes in clinical contexts – for example, in depression or anxiety disorders. The model opens up new perspectives in health research in particular – for example, by helping us understand how people with different psychological conditions make decisions. The dataset is set to be expanded to include demographic and psychological characteristics.
Centaur: Bridging Theory and Prediction
Centaur bridges two previously separate domains: interpretable theories and predictive power. It can reveal where classical models fall short – and provide insights into how they might be improved. This opens up new possibilities for research and real-world applications, from medicine to environmental science and the social sciences.
“We’re just getting started and already seeing enormous potential,” says institute director Eric Schulz. Ensuring that such systems remain transparent and controllable is key, Binz adds – for example, by using open, locally hosted models that safeguard full data sovereignty.
Next, the researchers aim to take a closer look inside Centaur: Which computational patterns correspond to specific decision-making processes? Can they be used to infer how people process information – or how decision strategies differ between healthy individuals and those with mental health conditions?
The researchers are convinced: “These models have the potential to fundamentally deepen our understanding of human cognition – provided we use them responsibly.” That this research is taking place at Helmholtz Munich rather than in the development departments of major tech companies is no coincidence. “We combine AI research with psychological theory – and with a clear ethical commitment,” says Binz. “In a public research environment, we have the freedom to pursue fundamental cognitive questions that are often not the focus in industry.”
What is Psych-101?
Psych-101 is a dataset specifically compiled by the team led by Marcel Binz for training the Centaur AI model. It contains over ten million individual decisions made by more than 60,000 participants across 160 psychological experiments. These experiments cover a wide range of human behavior, from risk-taking and reward learning to moral dilemmas. The researchers manually processed and standardized all the data to ensure that it could be interpreted by a language model. As such, Psych-101 represents a unique resource for systematically modeling human behavior based on natural language inputs.
Reference: “A foundation model to predict and capture human cognition” by Marcel Binz, Elif Akata, Matthias Bethge, Franziska Brändle, Fred Callaway, Julian Coda-Forno, Peter Dayan, Can Demircan, Maria K. Eckstein, Noémi Éltető, Thomas L. Griffiths, Susanne Haridi, Akshay K. Jagadish, Li Ji-An, Alexander Kipnis, Sreejan Kumar, Tobias Ludwig, Marvin Mathony, Marcelo Mattar, Alireza Modirshanechi, Surabhi S. Nath, Joshua C. Peterson, Milena Rmus, Evan M. Russek, Tankred Saanum, Johannes A. Schubert, Luca M. Schulze Buschoff, Nishad Singhi, Xin Sui, Mirko Thalmann, Fabian J. Theis, Vuong Truong, Vishaal Udandarao, Konstantinos Voudouris, Robert Wilson, Kristin Witte, Shuchen Wu, Dirk U. Wulff, Huadong Xiong and Eric Schulz, 2 July 2025, Nature.
DOI: 10.1038/s41586-025-09215-4
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
10 Comments
Do I let a trolley continue on and run into a group of children playing on the tracks or switch it to divert it to a track where my spouse is working and will likely get killed?
Well, I am not a psych student. I am an engineer. I set the switch half way, the trolley leaves the tracks and rather quickly comes to a halt without power. I like the Kobayashi Maru solutions. I wonder how the AI would do with that sort of prediction.
{o.o}
It would choose the solution that kills the fewest people. But they still get run over.
What a horrible idea.
We are going to live a long life if you respect your plastic brain, glia, excersize. Superfoods. Future looks good just get rid of D J Trump
President Trump’s term will expire in about 3 1/2 years. Suggesting getting “rid” of him borders on treason and insurrection. In other words, a rejection of the Rule of Law, as defined by the Constitution of our Republic. Perhaps you would care to expand on just how and why you would legally “get rid” of Trump. I don’t think that you have thought it through. The point of democratically electing certain offices it to avoid violence. Whether you are personally happy with the results or not, the greater good that comes from elections is avoiding violence. Those who whine about the results of a fair election are basically saying they only support democracy when the results go the way they want. Those who act as obstructionists to a duly elected officer of the electorate are basically hypocrites.
What you say is indeed correct. Oddly enough a President of the USA may be impeached legally. Your Republic is founded on the experiences of Britain where an obnoxious head of state ended up getting his head chopped off for demanding the Divine Right to do what he wanted. That demand led to a 7-year long civil war which cased the deaths of upwards of 250 000 people according to an Americanhistorian. And led to the Cromwellian puritannical dictatorship, which was not repeated. Which led to appointment of a new King, who was allowed to be head of the armed forces, but Parliament controlled the budget for the armed forces and, if I recall rightly, had to approve any declaration of war by the King.
It seems that a USAian President may unilaterally break treaties with other nations and unilaterally order the dropping of bombs on people without a declaration of war being declared by Congress, which would appear to abrogate the common-sense approach to division of power developed during the 1660s in the UK.
I advocate for The Rule of Law. As you point out, the mechanism for impeachment exists. IF Trump has committed impeachable offenses, then Congress should act. However, in our representational form of government, it isn’t sufficient that an individual citizen, and especially NOT a non-citizen like yourself, to be convinced that impeachment should take place. It is up to Congress to act.
Congressional war powers are an anachronism from the days that it took weeks to cross the oceans with troops. Since the invention of aircraft, and later ICBMs that can hit their targets in 20 minutes, it is obvious that requiring a body of hundreds of legislators to convene and discuss the need for action could result in annihilation before finishing a roll call. Technology has its down sides! We no longer have the luxury of considered debate before acting to protect ourselves.
The best information available suggests that that the world was within weeks of Iran being able to directly, and by their actions induce Israel to retaliate similarly, create radioactive fallout that would drift around the world, especially Europe and Asia.
I should point out also that what is generally called the Korean War, which the UK and other nations participated in, was ‘officially’ a UN police action. There is, therefore, a long-established precedent of not just the USA, but other countries, engaging in hostilities without a formal declaration of war. The USA, being the only country with Bunker-Busting bombs, would have been derelict in its responsibilities to its citizens and the world to NOT try to prevent Iran from making nuclear weapons. While there were probably people killed as collateral damage in the attacks on the U-enrichment sites, the attacks were restricted to the sites that created an imminent threat to the whole world. You have an unrealistic expectation, or a naive view of the danger of radioactive fallout, if you think that Trump should have waited until Iran proved it had nuclear weapons and Congress debated declaring war.
A goal of emulating human decisions may not be a good idea. Many, if not most, human decisions are irrational or, at best, non-rational. Emotion, located in the reptilian part of the brain, plays an important role in human behavior. Look at how it shapes the thinking of ‘Glia guy’ above. What is the advantage of using machines if they have the same potential flaw as humans? Making an illogical decision faster than a human does not strike me as being progress.
Who believes you have to try to be happy? I ask because we live in the only country that promoted freewill. Regardless of the one the people chose to represent it.
every time i see a new advancement in AI i think about how many people haven’t yet seen or played detroit become human