Accuracy of Manual Spike Sorting: Results for the Utah Intracortical Array

F. Wood1; M. Fellows2; C. Vargas-Irwin2; M.J. Black1*; J.P. Donoghue2

1. Dept. of Computer Science,
2. Neuroscience,
Brown Univ, Providence, RI, USA

Most Studies Of Neural Coding Use Manually Sorted Spike Trains Despite Evidence Of Subjective Variability Among Human Spike Sorters. This Variability Has Been Shown To Be Significant For Recordings From Tetrodes And Single Electrodes [1], Where Absolute False Positive (Fp) And False Negative (Fn) Error Rates Were Computed By Comparing Sorted Extra-Cellular Recordings To Intra-Cellular Ground Truth. The Use Of Utah Arrays Effectively Prohibits Individual Electrode Placement And Simultaneous Intra-Cellular Recording. In Lieu Of Intra-Cellular Recordings, We Constructed Synthetic Data For Which We Knew The Ground Truth. We Demonstrated That Expert Sorters, In A Two-Alternative Forced-Choice Experiment, Could Not Differentiate Between Synthetic And Real Data. A Combination Of Both Real And Synthetic Data Was Manually Sorted By Several Experts Using Commercial Software [3]. We Found Large Discrepancies In The Number Of Units Identified And The Spikes Assigned To Each By Different Experts. The Magnitude Of These Discrepancies Was Statistically Similar For The Real And Synthetic Data. We Calculated 23% Fp And 30% Fn Error Rates For The Synthetic Data And Postulate That These Rates Are Similar For Real Data. Our Results Further Suggest That Models Of Neural Coding Could Benefit From The Development Of New Statistical Techniques That Account For The Inherent Ambiguity Of Spikes.

[1] Harris, K., Henze, D., Csicsvari, J., Hirase, H., Buzsaki, G. (2000). Accuracy of tetrode spike separation as determined by simultaneous intracellular and extracellular measurements. J. Neurophys. 84(1):401-14.

[2] Maynard, E., Nordhausen, C., Normann, R. (1997). The Utah intracortical electrode array: A recording structure for potential brain-computer interfaces. EEG Clin. Neurophys. 102(3):228-39.

[3] Plexon, Inc., OffLineSorter, Dallas, TX

Support Contributed By: NIH-NINDS #NS25074, NIH-NINDS #N01-NS-2-2345, NSF ITR #0113679


Accuracy of manual spike sorting: Results for the Utah intracortical array,
Wood, F., Fellows, M., Vargas-Irwin, C., Black, M. J., Donoghue, J., P.,
Program No. 279.2. 2003, Abstract Viewer and Itinerary Planner, Washington, DC: Society for Neuroscience, 2003. Online.

Related Publications

On the variability of manual spike sorting,
Wood, F., Black, M. J., Vargas-Irwin, C., Fellows, M., Donoghue, J. P.,
submitted: IEEE Trans. Biomedical Engineering.