Robust Statistical Methodology in Detecting Irregular Firing Pattern in Dopaminergic Neurons

• Activity Number: 351
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Mental Health Statistics Section
• Author(s): Sudip Roy*+ and Daijin Ko
• Companies: University of Texas at San Antonio and University of Texas at San Antonio
• Keywords: Dopaminergic neurons ; Spike train ; Qn estimator ; Threshold ; SLISI

Abstract

Midbrain dopaminergic neurons in vivo have different firing patterns. They have different firing a phasic burst when reward exceeds prediction, or pause when an expected reward does not occur. Proposed a method, robust way of detecting additive patches outliers considering them as bursts or pauses from contaminated autoregressive time series model of the neuron spike train, inter spike interval (ISI) length. The location parameter (L) as robust Huber M-estimator and scale as robust Qn scale estimator to standardize the log transformed ISIs (SLISI). The robust serial correlation and neurophysiology literature show autoregressive nature of spike trains. Robust AR regression model with high breakdown point (Gervini and Yohai, 2002) fitted. Then residuals are analyzed probabilistically. An automated data adaptive method to identify outliers is developed based on the distribution of the standardized residuals and detect extended clusters of bursts and pauses. A threshold with adjustment factor been introduced which efficiently can discriminate outliers from extreme values of the distribution.