Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

An Electronic Phenotype Series for Measuring Adherence to the Wake Up and Breathe Protocol in Mechanically Ventilated ICU Patients.

Research square·2026

Same author

Auditory Cortex Distinguishes between Spontaneous and Sound-Evoked Movements.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2026

Same author

Hierarchical recurrent temporal prediction as a model of the mammalian dorsal visual pathway.

PLoS computational biology·2026

Same author

Intelligent Reasoning Cues: A Framework and Case Study of the Roles of AI Information in Complex Decisions.

Proceedings of the SIGCHI conference on human factors in computing systems. CHI Conference·2026

Same author

Rethinking hierarchy: the auditory system as an integrated cortical-subcortical network.

Nature reviews. Neuroscience·2026

Same author

Dominant baboons experience more interrupted and less rest at night.

Current biology : CB·2025

Search research articles

Related Experiment Video

Updated: Mar 25, 2026

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

Measuring the Performance of Neural Models.

Oliver Schoppe¹, Nicol S Harper², Ben D B Willmore²

¹Department of Physiology, Anatomy, and Genetics, University of OxfordOxford, UK; Bio-Inspired Information Processing, Technische Universität MünchenGarching, Germany.

Frontiers in Computational Neuroscience

|February 24, 2016

Summary

Evaluating neural models requires metrics that account for inherent response variability. A new method accurately calculates the normalized correlation coefficient (CC norm), making it a superior choice over Signal Power Explained (SPE) for assessing model performance.

Keywords:

model selection neural coding receptive field sensory neuron signal power statistical modeling

More Related Videos

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Related Experiment Videos

Last Updated: Mar 25, 2026

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

Published on: March 2, 2015

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model

Published on: April 18, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Area of Science:

Computational neuroscience
Statistical modeling
Sensory neurophysiology

Background:

Accurate performance metrics are crucial for comparing statistical and computational models, especially for predicting neural responses to sensory stimuli.
Neural responses exhibit inherent variability, complicating standard performance metrics like the correlation coefficient by conflating explainable and unexplainable variance.
Existing metrics, Signal Power Explained (SPE) and normalized correlation coefficient (CC norm), attempt to address this variability but have limitations.

Purpose of the Study:

To analyze and compare the performance metrics SPE and CC norm for neural models.
To address the limitations of SPE, specifically its unbounded nature and potential for negative values.
To develop a method for direct, accurate, and efficient calculation of CC norm.

Main Methods:

Analysis of existing metrics SPE and CC norm, demonstrating their mathematical relationship.
Identification of a novel computational approach for calculating CC norm.
Comparison of the practical utility and interpretability of SPE and CC norm.

Main Results:

SPE is shown to be unbounded and can produce negative values, making interpretation difficult.
CC norm is bounded between -1 and 1, offering more straightforward interpretation.
A new, accurate, and efficient method for calculating CC norm has been developed.

Conclusions:

The normalized correlation coefficient (CC norm) is a more robust and interpretable metric for evaluating neural models than SPE.
The newly developed method for calculating CC norm enhances its practical applicability.
This advancement facilitates more reliable comparison and selection of computational models for neural responses.