1. Home
2. Research Result
3. Press Release

ADVANCED TELECOMMUNICATIONS RESEARCH INSTITUTE INTERNATIONAL - JAPAN
UNIVERSITY OF CALIFORNIA LOS ANGELES - USA

Is it possible to directly boost one’s own confidence by directly training the brain?
Researchers have discovered a way to implicitly amplify confidence in the brain by combining the use of artificial intelligence and brain imaging technology. The study, published in the journal Nature Communications, could lead to important applications in clinical, medical and social settings.
Self-confidence is an essential quality to succeed in the world, such as in business environments, politics or many other aspects of our everyday life. Furthermore, confidence is an important aspect in mental illnesses such as depression and Alzheimer’s disease, where the condition is often further complicated by patients thinking negatively of their own capacities. Recent advances in neuroscience have highlighted the plasticity of the brain, indicating it is malleable even later in life.
The international team developed a state-of-the-art method to read and then amplify a high confidence state using a new technique called ‘Decoded Neurofeedback’. This technique used brain scanning to monitor and detect the occurrence of specific complex patterns of activity corresponding to high confidence states, while participants performed a simple perceptual task. In the training sessions, whenever the pattern of high confidence was detected, participants received a small monetary reward. This experiment allowed researchers to directly boost one’s own confidence unconsciously, i.e. participants were unaware that such manipulation took place. Importantly, the effect could be reversed, as confidence could also be decreased.
Dr. Mitsuo Kawato, Director of the Computational Neuroscience Laboratories at ATR, Kyoto, and one of the authors on the study, has pioneered this state-of-the-art technology. He explained the process:
“How is confidence represented in the brain? Although this is a very complex question, we used approaches drawn from artificial intelligence (AI) to find specific patterns in the brain that could reliably tell us when a participant was in a high or low confidence state. The core challenge was then to use this information in real-time, to make the occurrence of a confident state more likely to happen in the future”.
Dr. Aurelio Cortese, of the Advanced Telecommunications Research Institute International, Kyoto, led the research:
“Surprisingly, by continuously pairing the occurrence of the highly confident state with a reward - a small amount of money - in real-time, we were able to do just that: when participants had to rate their confidence in the perceptual task at the end of the training, their were consistently more confident”.
Dr. Hakwan Lau, Associate Professor in the UCLA Psychology Department, was the senior author on the study and an expert in confidence and metacognition:
“Crucially, in this study confidence was measured quantitatively via rigorous psychophysics, making sure the effects were not just a change of mood or simple reporting strategy. Such changes in confidence took place even though the participants performed the relevant task at the same performance level”.
The sample size was relatively small (17 people), but is in line with basic science investigations of similar kinds. The team is currently working on the development of potential new clinical treatment for patients with various psychiatric conditions.
The Original Paper: Aurelio Cortese, Kaoru Amano, Ai Koizumi, Mitsuo Kawato, Hakwan Lau: Multivoxel neurofeedback selectively modulates confidence without changing perceptual performance. Nature Communications.
DOI: 10.1038/ncomms13669

The team, in a different study led by co-author Dr. Ai Koizumi, has indeed already discovered a new way to unconsciously erase fear memories, reprogramming the brain to overcome fear. The study was recently published in the inaugural edition of Nature Human Behavior on November 21st, and opens the potential for radical new treatments of conditions such as post-traumatic-stress-disorder (PTSD) and phobias.


The study was conducted in ‘Brain Machine Interface Development’ under the Strategic Research Program for Brain Sciences by AMED of Japan. A.C. was supported by a Japanese Government Monbukagakusho MEXT grant. This work was supported partially by the National Institute of Neurological Disorders and Stroke of the National Institutes of Health (Grant No. R01NS088628) and the Templeton Foundation (Grant No. 21569), to H.L.