Blind trust in enhancement technologies encourages risk-taking even if the tech is a sham

A placebo effect can make users overconfident when they think tech is helping them
Illustration of a white-to-light-blue side profile of a human face with network nodes pointing out against a blue background
The research article was published in "Computers in Human Behavior". Illustration: Feodora Chiosea/iStock

A new study suggests that a placebo effect is at play when people expect their performance to be enhanced by augmentation technologies, such as artificial intelligence (AI). The researchers found that individuals with high expectations of these technologies engage in riskier decision-making, which could be a problem as people adopt these technologies without properly understanding their benefits and limits.

Augmentation technologies boosting our physical, cognitive, or sensory performance have become commonplace. Some are so widely in use that they’ve become invisible – spellcheck, for example – and new technologies are emerging that could push our abilities beyond human limits, like exoskeletons and AI-based vision-enhancement. But the hype around these technologies also builds expectations, which could lead people to change their behaviour.

‘Individuals are more inclined to take risks when they believe they are enhanced by cutting-edge technologies like AI or brain-computer interfaces,’ says Robin Welsch, assistant professor at Aalto University. ‘This occurs even if no actual enhancement technology is involved, indicating that it’s about people’s expectations rather than any noticeable improvement. The findings also imply that a strong belief in improvement, based on a fake system, can alter decision-making.’

Don’t trust the processor

Together with colleagues at LMU Munich, HU Berlin and Aalto University, Welsch measured decision-making and risk-taking behaviour with a well-known psychological experiment, the Columbia Card Task. In the experiment, participants win or lose points by turning over cards with hidden values. The 27 participants were led to believe that an AI-controlled brain-computer interface, the placebo, would enhance their cognitive abilities by using binaural sounds to track the loss cards. 

But the game was rigged – the augmentation provided no real benefit, and participants almost never encountered a loss card. Still, most of the participants thought the augmentation had helped them do better, and this made them take on greater risks. These findings show how sham cognitive enhancements can have real effects on risk-taking.

‘The hype surrounding these technologies skews people's expectations,’ says Steeven Villa, doctoral researcher at LMU Munich. ‘It can lead people to make riskier decisions and favourable user evaluations, which can have real consequences.’

Male participant in psychological study with sensors attached to his head clicking through an experiment on a computer in a white office space
The researchers measured risk-taking was with the Columbia Card Task experiment. Image: Robin Welsch

Technology is never neutral

Modern AI-based vision enhancement tools are becoming available for firefighters to see through smoke, and factory workers are already taking advantage of exoskeletons to handle loads that defy human physique. The creeping effects of generative AI systems could offer similar boosts to knowledge workers. The authors are worried that entire professions could develop a larger appetite for risks, built on misguided expectations.

‘AI-based technologies that enhance users are increasingly common and play a role in real-life decisions that impact people's lives, well-being, confidence, and safety.’ says Thomas Kosch, professor at HU Berlin. ‘To ensure the effectiveness of new technologies beyond the hype, placebo-controlled studies are necessary for accurate evaluation and validation to tell apart snake-oil from real innovation.’

Reference: Steeven Villa, Thomas Kosch, Felix Grelka, Albrecht Schmidt, Robin Welsch (2023). The placebo effect of human augmentation: Anticipating cognitive augmentation increases risk-taking behavior. In Computers in Human Behavior, Volume 146. doi:

Engineering psychology students Salla Nicholls and Esko Evtyukov, photo by Matti Ahlgren

Salla Nicholls and Esko Evtyukov are thrilled to be the first engineering psychology students

Esko Evtyukov hopes to work in the games industry in the future and is currently working as a research assistant in Assistant Professor Robin Welsch's team, designing virtual reality experiments. Salla Nicholls, by contrast, dreams of combining data science and human-computer interaction.

Artistic illustration for engineering psychology bachelors programme in Aalto University. Human face, laptop, game controller and phone. Illustration by Matti Ahlgren

Engineering Psychology

Understanding humans and technology

Department of Computer Science
illustration of a green chat bubble against a yello background with yellow round objects in the middle portraying a "person is writing" prompt.

Chat AIs can role-play humans in surveys and pilot studies

Synthetic data from large language models can mimic human responses in interviews and questionnaires. Research data from popular crowdsourcing platforms may now contain fake responses that cannot be reliably detected, raising the risk of poisoned data

Assistant Professor Robin Welsch sitting on a sofa against a green backdrop

Psyched up for computer science

How can we align technology with the way we think about, feel, and understand the world? Robin Welsch is a psychologist-turned-computer scientist on a quest to find an answer.

  • Published:
  • Updated:

Read more news

Photo: Tima Miroschnichenko, Pexels.
Press releases Published:

In low-hierarchy organisations, even key policy issues are discussed in Slack

In a recent study, Aalto University alumn Lauri Pietinalho, a visiting scholar at New York University's Stern School of Business, and Frank Martela, an assistant professor at Aalto University, investigated how low-hierarchy organisations deal with shared policies in confrontational situations and how authority functions within them.
bakteereja ohjataan magneettikentän avulla
Press releases, Research & Art Published:

Getting bacteria into line

Physicists use magnetic fields to manipulate bacterial behaviour
border crossings 2020
Press releases, Research & Art Published:

Nordic researchers develop predictive model for cross-border COVID spread

The uniquely multinational and cross-disciplinary research was made possible by transparent data-sharing between Nordic countries.
Artistic illustration: Algorithms over a computer chip
Research & Art Published:

Aalto computer scientists in ICALP 2024

Five papers from the Department of Computer Science were accepted to International Colloquium on Automata, Languages, and Programming (ICALP).