The Role of LGN Bursts During Natural Vision

Nicholas A. Lesica and Garrett B. Stanley
Division of Engineering and Applied Sciences
Harvard University
The role of LGN burst responses during visual encoding remains an open question. Here, we characterize the function of LGN bursts in signaling salient features of natural stimuli. Burst events were reliably triggered by specific stimulus events, such as the movement of an object into the receptive field. A significant increase in bursting was observed during natural stimulation (relative to white-noise stimulation) and linked to the strong correlation structure of the natural stimulus. To characterize the role of LGN bursts in the neural code, responses to natural scene movies were predicted using a linear model (with gain control and rectification) and compared to experimentally observed responses. While the linear model successfully predicted LGN responses to natural scene movies during tonic firing, it typically underestimated the response during periods of bursting. Thus, LGN bursts provide a nonlinear amplification of stimulus features that are characteristic of correlated natural stimuli. To account for this nonlinearity, an encoding model was developed in which the firing state of the neuron (burst or tonic) is determined by the time since the last spike. The parameters of the model were optimized so that the stimulus related information carried by burst and tonic spikes approximated that observed experimentally, indicating that LGN encoding of natural scenes is well described by the model. The results of this study highlight the importance of LGN burst events in the encoding of natural stimuli and transmission of visual information to downstream neurons.