Neural decoding of semantic concepts: a systematic literature review

Area of Science:

• Neuroscience
• Neural Engineering
• Cognitive Science

Background:

• Semantic concepts are fundamental to human cognition and understanding.
• Neuroscience is advancing in mapping neural activity patterns for specific concepts.
• Semantic decoding aims to identify a person's current thoughts from their neural activity.

Purpose of the Study:

• To systematically review the literature on semantic neural decoding across various neuroimaging modalities.
• To quantify the efficacy of current semantic decoding methods.
• To identify challenges and future research directions in semantic neural decoding.

Main Methods:

• Systematic literature review following PRISMA guidelines.
• Searched PubMed and Google Scholar for relevant peer-reviewed studies.
• Included 74 studies that used semantic neural decoding to identify concepts from neural activity.

Main Results:

• Modern neuroscientific tools enable decoding of individual concepts using diverse neuroimaging modalities.
• Specific neuroimaging techniques, experimental designs, and machine learning pipelines are utilized.
• Information transfer rates were used to quantify semantic decoder efficacy.

Conclusions:

• Semantic decoding is a rapidly advancing research area with significant potential.
• This review provides a comprehensive overview and efficacy quantification, addressing a gap in existing literature.
• Challenges and future directions, including speculative ones, are discussed to guide further research.