Confounding of norm-based and adaptation effects in brain responses
David Alexander Kahn1, Geoffrey Karl Aguirre
1Department of Neurology, University of Pennsylvania, 3 West Gates, 3400 Spruce Street, Philadelphia, PA 19104, United States.
Neuroimage
|March 8, 2012
Summary
Neural coding studies can confuse adaptation and prototype effects due to experimental design. This research demonstrates the confound and how to minimize it for accurate perceptual representation research.
Area of Science:
- Neuroscience
- Cognitive Science
- Psychology
Background:
- Neural coding research investigates how the brain represents perceptual stimuli.
- Two key properties studied are adaptation (recovery from adaptation) and prototype effects (response to distinct stimuli).
- Standard neuroscience paradigms often confound these two effects, leading to misinterpretation of results.
Purpose of the Study:
- To identify and demonstrate the confounding of neural adaptation and prototype effects in perceptual coding studies.
- To simulate Blood-Oxygen-Level-Dependent (BOLD) functional magnetic resonance imaging (fMRI) results under each effect.
- To propose methods for minimizing this intrinsic confound in experimental design.
Main Methods:
- Simulated BOLD fMRI data to model the impact of adaptation and prototype effects.
- Designed an example fMRI experiment to illustrate the confound.
- Analyzed how stimulus dissimilarity and proximity to stimulus space centers influence neural responses.
Main Results:
- Simulations showed that each effect can produce spurious results supporting the other.
- Apparent neural similarity effects can arise from underlying prototype representations.
- Apparent prototype effects can result from differential adaptation.
Conclusions:
- A fundamental confound exists between adaptation and prototype effects in neural coding research.
- This confound can lead to incorrect conclusions about perceptual representation and category learning.
- Careful experimental design is crucial to disentangle these effects and accurately study neural coding.
