Our human need to handle with numerous visual information in daily life.Due to the limitation of brain resources,only the key part of visual information will pass the screening,be processed by the brain,and finally guide the behavioral decision-making,which is the mechanism of attention.Attentional modulation may mainly originate from the frontal and parietal cortex,and has an impact on almost all visual cortex regions through extensive neural feedback.The primary visual cortex is an early stage of visual information processing in the brain,and the physiological structure of V1 region,neuronal response property and non-classical receptive field has been thoroughly studied.The evoked visual response of a neuron is the result of the combined influence of a larger range of visual information from inside and outside its classical receptive field.With the development of technology in recent years,such as f MRI and multichannel neuro-electrophysiological recording,more and more researchers have found that the V1 area is also modulated by visual attention.Moreover,attention can change the responses of individual neurons to visual stimuli.This modulation is related to many factors,such as the relative position of the attention focus and the receptive field of the neuron,and the amount of the attentional load.It is reported that spatial attention can significantly affect the correlated activity between neural populations.However,how spatial attention modulates the interactions between neurons in V1 region are still unclear.Aiming to address the problem,this study adopts the adjustable microelectrode array technique to examine the modulation of spatial attention on the spike count correlation and Granger causality between spiking responses of neurons with non-overlapping receptive fields in macaque primary visual cortex.We explored the effects when attentional focus shifts and the influence of task difficulty on the spike count correlations,and discussed the underlying neural mechanisms.What kind of impact does attention pose onto the horizontal function connections between neurons with non-overlapping receptive fields in the primary visual cortex,and whether the influence will be the same among different types of neuronal pairs etc.? The answers to the series of questions above are of great significance to understand the neural information processing mechanisms of attention.We simultaneously recorded spiking responses from pairs of nonoverlapping receptive field neurons in the primary visual cortex of one hemisphere of the rhesus monkey while the animal was awake and performing a cued attentional task.Using spike count correlations and granger causality among pair of neurons,we systematically studied the influence of attention focus on response correlation between neurons.The results showed that the mean of the spike count correlation among pairwise neurons decreased with the attention focus moving away from their receptive fields,whereas the Granger causality increased instead.This significant modulation was only founded between simple and complex cells in superficial layer of V1,which received feedback input from higher cortical areas.The phenomenon suggested that it might be due to attention-related influence.In addition,we found that the preferred stimulus direction of neurons was also associated to the attentional modulation,but not the distance between neurons.We also examined the effect of increased attentional load by elevating task difficulty on the spike count correlations between pairs of neurons.We first classified the neurons into groups as enhanced neurons(E),suppressed neurons(S)and not significantly modulated neurons(N),which was based on whether their spiking responses were significantly modulated by attention during hard task.Analysis was mainly applied on three types of neural pairs in this study:EN,SN and NN.Moreover,we used multiplicative attentional ratio of firing rate to quantitively examine the relationship between the extent of competition among neural pairs and task difficulty.Our results showed that spike count correlations between V1 neurons decrease in general during hard task.However,different types of neural pairs exhibited distinct patterns.Correlations among EN pairs shifted to negative when they were more competitive,and moderately decreased when the task was difficult.Difficulty significantly decreased the correlations among NN pairs,but no tendency was found between correlations and neural competition.Instead,correlations among SN pairs would remarkably increase when the task was hard,and slowly rise along with the extent of competition.The results indicated that:(1)Although the spike count correlation between most of V1 neurons decreased with higher task difficulty,some of theirs increased instead,which was directly related to the attentional modulation of neurons;(2)Biased competition was only one of many types of neuron interactions,and the integration of visual information was likely to be realized through the complex interactions between neurons.Our work utilized methods of psychophysics,multi microelectrode array as well as in vivo neural electrophysiological recording technique to investigate the influence of different attended locations and attentional load on the horizontal functional connectivity between V1 neurons whose receptive fields were non-overlapping.It helped to commprehend underlying neural mechanisms of how the attention filtering information and to improve the cognitive level,provided important experimental basis for understanding the influence of attention load on response correlations between pairs of neruons and for neural network modeling of the visual system of the brain. |