Can The Trends Cogn Sci machine change our minds?

If we can provide the machine with information about our brain activity, can the machine use that information to prevent some bad mental state from occurring? For example, can the machine learn how to relieve subjective pain? In a recent experiment, the University of Cambridge designed a closed-loop brain-machine interface that learned to reduce participants' pain by decoding pain-related brain activity.
doing so, they also highlighted some of the challenges associated with adaptive processes in brain-computer communication.
In the study, Suyi Zhang and colleagues, instead of using sensory feedback, implemented an enhanced learning algorithm that dynamically adjusted the interaction between the machine and the participants, choosing the most effective measures to reduce the painful electrical stimulation suffered by the participants.
, and most importantly, this particular setting also gives participants the ability to optimize communication with the machine by voluntarily altering their brain response to pain.
this raises an interesting and problematic problem of co-adaptation: machines not only learn to adjust their movements according to their participants' reactions, but also learn to communicate with the machine by regulating their brain activity.
to explore the potential challenges associated with adaptation, Suyi Zhang and colleagues designed experiments that conceptually mimic an algorithm that learns to optimize the parameters of pain-stimulating treatment.
However, the brain-computer interface has learned to choose between a low-intensity shock from two electric stimulators or a high-intensity shock, and the machine can quickly learn how to adjust its movements on a trial-by-test basis, thereby gradually reducing the number of painful electric shocks.
that adaptive communication between the brain and the computer is feasible.
The second part of the study was when participants were trained to predict high and low shocks of passive prediction to predict brain activity to predict pain prediction patterns, but between the first training and subsequent interactive training, not only did the prediction accuracy decrease significantly, but subsequent searchlight analysis also confirmed that there was less information about pain prediction in the brain.
that the only behavior that attempts to communicate with a machine interferes with its actual ability to communicate with it.
they speculate that the fact that they are only trying to communicate with a brain decoder seems to affect the treatment of pain, which in turn may regulate the decoding of pain in the brain.
In fact, a great deal of research is still needed, which paves the way for broader and deeper research in the future, and future studies may also want to consider training target decoders with larger data sets to improve decoding accuracy and expand the scope of potential interventions."
this, consider using feature alignment methods, such as over-alignment or shared response models, because they have previously proven useful in other real-time decoding methods.
If clarifying the complex interactions between the brain and machines may prepare for new interventions in cognitive neuroscience, much work needs to be done before machines can flexibly monitor and change abnormal brain patterns for therapeutic purposes, although the Cambridge University study suggests that this is a possibility worth considering at this time."
: Vincent Taschereau-Dumouchel, Mathieu Roy, et al. Can Brain Code Machines Change Our Minds? Trends in cognitive sciences. DOI: Zhang Ao Source: MedSci Original Copyright Notice: All text, images and audio and video materials on this website that indicate "Source: Mets Medicine" or "Source: MedSci Originals" are owned by Mets Medicine and are not authorized to be reproduced by any media, website or individual, and are authorized to be reproduced with the words "Source: Mets Medicine".
all reprinted articles on this website are for the purpose of transmitting more information and clearly indicate the source and author, and media or individuals who do not wish to be reproduced may contact us and we will delete them immediately.
at the same time reproduced content does not represent the position of this site.
leave a message here