Neural mind reading of multi-dimensional decisions by monkey mid-brain activity

Area of Science:

• Neuroscience
• Biomedical Engineering
• Cognitive Science

Background:

• Brain-machine interfaces (BMIs) offer significant potential for enhancing the quality of life for individuals with disabilities.
• Developing effective cognitive BMIs requires accurate methods for decoding neural signals related to decision-making processes.

Purpose of the Study:

• To develop and validate neural mind-reading techniques for cognitive BMIs.
• To provide a real-time readout of decision processes using neural activity.

Main Methods:

• Trained two monkeys on go/no-go tasks.
• Monitored neuronal activity in the mid-brain superior colliculus (SC).
• Designed a virtual decision function (VDF) to analyze ensemble SC neuron activity for predicting decisions.

Main Results:

• The VDF successfully predicted cue location and the monkeys' motor choices (go/no-go) shortly after cue presentation.
• Demonstrated the ability to decode continuous decision progress from neural activity on a single-trial basis.

Conclusions:

• The developed neural mind-reading techniques show promise for rapid, real-time control of communication support devices.
• These findings advance the potential of cognitive BMIs for assistive technologies.