Beyond metaphors and semantics: A framework for causal inference in neuroscience

Area of Science:

• Neuroscience
• Cognitive Science
• Philosophy of Science

Background:

• The prevalent "coding" metaphor in neuroscience research is increasingly recognized as problematic.
• This metaphor often leads to the misinterpretation of correlational data as causal evidence.

Purpose of the Study:

• To critically evaluate the limitations of the coding metaphor in neuroscience.
• To advocate for a shift towards a more constructive syntactic approach for causal inference.

Main Methods:

• Conceptual analysis of the "coding" metaphor's implications in scientific literature.
• Review of existing research methodologies in neuroscience, focusing on correlational and conditional correlational designs.

Main Results:

• Correlational evidence, common in neuroscience, is frequently imbued with inappropriate causal power due to the coding metaphor.
• The aggregate of correlational studies does inform causal inference, but the metaphor hinders accurate interpretation.

Conclusions:

• Abandoning the problematic coding metaphor is crucial for advancing neuroscience.
• Developing a constructive syntax is essential for guiding principled causal inference from neuroscientific data.