RECEIVER OPERATING CHARACTERISTIC (ROC) ANALYSIS OF NEURAL CODE EFFICACIES .1. GRADED PHOTORECEPTOR POTENTIALS AND DATA QUALITY

Nerve cell signals are different in form from the stimuli that evoke t hem and they exhibit complex spatio-temporal characteristics. This def ines a neural coding problem which is addressed by two current theorie s: Multiple Meaning Theory holds that neural signals contain patterns that make statements about combinations of stimulus properties; the Ta sk Dependence Hypothesis suggests that different features of identical neural. signals mediate performance in different behavioral tasks. Th ese coding issues were addressed by investigating the representation o f sensory information in the distal nervous system after transduction of visual stimuli into bio-electric signals. The objects of study were light-evoked neural responses which had been intracellularly recorded from single retinula (photoreceptor) cells in Limulus lateral eyes. T he efficacies with which sensory information was represented by variou s candidate neural codes were calculated using receiver operating char acteristic (ROC) analyses to provide objective indices. The specific v isual problem under investigation was discrimination between light fla shes whose intensities differed by a very small amount. A wide range o f light adaptation states and relative stimulus intensities were explo red. Extremely stringent data quality standards were applied which res tricted the investigation to cells whose potentials did not exhibit an y statistically significant drift during the hours required for data c ollection. Seven cellular characterizations were simultaneously monito red to detect drift in a given cell's potentials; these characterizati ons included the value of the membrane potential and the values of six candidate codes. These codes were: the area under the light-evoked re ceptor potential (RP), the mean value of the RP, the peak height of th e RP, the slope of the onset of the RP, the duration required for the RP to drop from its peak by a given amount, and the duration required for the RP to end. The results were: (1) Light adaptation increases ef ficacy. (2) Thus, light adaptation trades sensitivity for acuity (as c haracterized by ROC discriminations). (3) Increasing relative light fl ash intensity also increases efficacy. (4) The efficacies of the vario us codes are significantly different and fail in the following order: area greater than or equal to peak = mean greater than or equal to dur ation-end = slope = duration-drop. These findings further demonstrate that arbitrary characterizations of stimulus-response relationships ar e very likely to be incomplete. They particularly indicate that many c ommonly used and quite conventional neural analysis strategies may sub stantially underestimate system performance.