首页 >> 中文网站 >> 新闻
                                                                                     脑网络组系列讲座 

                                                                                                                                                                                                                                                                  

                                                                                                                         [报  告]

  

  题    目  (TITLE)       :     Using two-photon calcium imaging to study macaque visual cortex

  讲座人  (SPEAKER):     余聪,教授,北京大学心理与认知科学学院

  主持人(CHAIR)   :     蒋田仔,研究员,中国科学院自动化研究所

  时   间   (TIME)        :      2023年4月25日,下午2:30-3:30

  地   点   (VENUE)    :     智能化大厦三层第三会议室

   

  

[摘  要]

 

    Two-photon calcium imaging allows simultaneous recording of thousands of neurons at cellular resolution. I will present three 2-p studies on macaque visual cortex, and discuss this technology from the perspective of a psychophysicist.

(1) Orientation, OD, and SF functional maps. In FOVs about the size of a hypercolumn, V1 superficiallayer neurons are clustered in orientation and OD preferences, but much less so in SF preferences that only cover a limited portion (2-3 octaves) of the visible SF range. The three maps do not intersect orthogonally or in parallel. Thus, hypercolumns may represent full ranges of orientation and OD, but not SF, preferences, and with no tight geometric restrictions. 

(2) Plaid detectors. Psychophysicists have long speculated that V1 neurons can detect two-dimensional plaid patterns as textons, but neuronal responses to plaids are typically suppressed (cross-orientation suppression). Still, 2-p imaging revealed a subgroup of V1 neurons that prefer plaid patterns to Gabor gratings, with plaid orientation preferences. These plaid detectors are likely overlooked by electrode recordings that use gratings/lines to map RFs. 

(3) Texture segregation. For texture-based figure-ground segregation, whether V1 can detect the figure, or only the figure-ground texture border, remains debated, while V4 is agreed to play a major role. We applied PCA to reduce the dimension of V1 and V4 responses to figure-ground texture stimuli, and then used linear SVMs to calculate the decoding accuracies for border detection and figure-ground segregation.

   

[个人简介] 

    

   余聪,北京大学心理学院教授,国家杰出青年基金获得者。余老师实验室运用心理物理方法研究人类视觉,尤其是知觉学习的心理物理与脑机制以及知觉学习在临床和工程中的应用。近年,他运用清醒猴双光子钙成像技术,研究视觉皮层神经元的群体反应特性以及视知觉过程的神经机制。

附件: