Study On Coding Mechanism Of Fear Information In SC Of Rats During Cognitive Learning Of Natural Images

The emotion of fear is crucial to the survival of animals,and vision is the most important way for animals to obtain fear information.In the process of processing fear visual information,the superior colliculus(SC)plays a very important role,so it is of great significance to explore the response mechanism of SC region to fear information.In previous studies on fear vision in rats,simple images were mostly used as visual stimuli.Compared with simple images(e.g.,"?","?",etc.),natural images(e.g.,"snake","tree",etc.)have richer details and more varied shapes.However,when complex natural images are used as visual stimuli,the coding mechanism of the fear information brought by the complex natural images in the SC area of the rat and the coding effects of the SC area on the different characteristics of the natural images are still not very clear.In order to explore these problems,in this paper,the Long Evans(LE)rats were used as experimental animal,and a visual cognitive training experiment scheme based on three natural images with different emotional attributes was designed.Through the analysis of Spike signals and local field potential(LFP)signals,the coding mechanism of fear information in the SC area of rats during the cognitive learning process of natural images was explored,and the coding effect of SC region on different features of natural image was analyzed.The specific research contents and results are as follows:(1)The conditioned fear association training method was adopted to design the experimental scheme based on three different emotional attributes of natural images(fear: "snake",neutral: "tree",reward: "rat food"),and the visual cognition training was carried out on the rats,and the effects of cognitive training on natural images in rats were analyzed.The results showed that the natural images of three different emotional attributes could be effectively recognized by LE rats after training.Afterwards,the neural response signals of neurons in the superior colliculus of LE rats to different stimulation images were collected by using the microelectrode array technology and Cerebus signal acquisition system.(2)First spike latency(FSL)and Spike firing rate were used to explore the coding mechanism of fear information in the SC region of LE rats,and finally,a classification model was constructed by using support vector machine(SVM)to verify the coding effect of the two methods.It was found that under the stimulation of the fear image,the FSL of the superior colliculus neurons decreased significantly,and the Spike firing rate increased significantly.The results showed that the SC region of the rats could be activated effectively by the fear information,and the information transfer rate of the SC neurons could be accelerated.(3)The ? frequency band related to fear information in LFP signals of rat superior colliculus neuron clusters was determined and extracted,then the fast Fourier transform and continuous wavelet transform were used to analyze the ? frequency band in time domain and time frequency domain,and finally the classification model was constructed by the support vector machine algorithm to verify the coding effect.It was found that under the stimulation of the fear image,the time-domain waveform of the ? frequency band oscillated significantly,and the time-frequency energy value increased significantly.The results showed that the fear information could be effectively characterized by the time domain waveform characteristics and time frequency energy characteristics of the ? band of LFP signals of the superior colliculus neurons cluster.(4)A multidimensional scale(MDS)method was used to analyze the neuron response of LE rat superior colliculus neuron clusters under various image stimulations,and the coding effect of SC region on different features of natural image was explored.It was found that the original and low-frequency images of the fear image could be effectively classified by the SC region,while other stimulus images could not be classified.The results showed that the original and low-frequency image features of the fear image could be encoded by the SC region of LE rats,while the two local image features of the fear image could not be encoded by the SC region.