Introduction
This corrigendum provides a formal correction to the 2013 Nature study, "Membrane potential dynamics of grid cells." While the primary scientific conclusions regarding the intracellular mechanisms of grid cell firing remain unchanged, a post-publication audit identified a technical calibration error in the experimental treadmill setup. This error resulted in an underestimation of the spatial distances traveled by the subjects in virtual reality (VR) environments.
Technical Calibration Error
In the original study, mice navigated one-dimensional linear tracks while head-fixed on a treadmill. Their movement was monitored via an analogue angular encoder. The conversion of this analogue rotation signal into digital distance data was found to be flawed, leading to reported track lengths that were 5% to 30% shorter than the actual physical distances traveled.
This discrepancy primarily affected the tetrode recording data used for identifying grid cell populations. Importantly, the whole-cell intracellular recordings—the core focus of the study's mechanistic insights—were unaffected by this specific scaling error.
Methodology for Correction
To address the error, the researchers re-processed the raw encoder signals to establish accurate track lengths. Because the tracks were longer than initially reported, the grid cell classifier required adjustment. This algorithm distinguishes grid cells based on the frequency of state transitions (firing vs. silent) over a set distance. The updated classifier was recalibrated to require three switches per 5.2 meters of track, reflecting the corrected spatial scale.
Updated Spatial Metrics and Results
The re-analysis of the tetrode VR data yielded updated quantitative values for several key spatial metrics. The actual track lengths for the recording sessions between July 2010 and November 2010 are detailed in the table below.
Following the correction, the revised spatial metrics are as follows:
- Mean Field Width: $40 \pm 11$ cm
- Peak Firing Rate: $12 \pm 6$ Hz
- Fields per Metre: $0.7 \pm 0.3$
- Track Lengths: Corrected values range from 411 cm to 1,035 cm.
Impact on Scientific Conclusions
Despite the shift in spatial scale, the fundamental findings of the paper are robust. The relationship between the animal's movement and the internal electrical state of the neurons remained consistent. The updated classifier identified the same set of neurons as grid cells, demonstrating that the spatial signatures of these cells are distinct enough to remain identifiable even with a 30% margin of error in distance measurement.
The authors confirm that these corrections have no impact on the results, conclusions, or interpretation of the original article. The discovery of how membrane potential dynamics support grid cell firing remains valid, and the record has been updated to ensure the accuracy of the spatial data for future research.
Summary updated