【作者】 俞黎平；

【导师】 孙心德；

【作者基本信息】 华东师范大学 ， 生理学， 2010， 博士

【摘要】 人和动物通过不同的感觉系统来接受和传递来自外界的各种信息,例如通过视觉系统感知和编码外界视觉信息,通过听觉系统感知和编码听觉信息等。然而,外界事件往往包含多种感觉模态信息,为能够真实地反映和编码外部事物,大脑需要对输入的信息进行整合。但是我们对神经系统信息整合还知之甚少。猫上丘为研究感觉信息的整合提供了理想模型。多种感觉模态信息(视觉,听觉和体感觉等)的集中输入,使上丘深层富含高比例的多感觉神经元。这些神经元在多感觉整合和感觉运动整合过程中发挥着重要作用,能够增强对外界信息的定位能力,加快对外界事物的反应能力。尽管经过30多年的研究,人们已经对多感觉系统的一些基本整合规则有了较为深入的了解,但是,对于一些基本问题我们仍然没有确切的答案。1.尽管多感觉整合能力的获得需要感觉经历,可是什么样的感觉经验对整合能力的发展是最重要的?例如一个听-视多感觉神经元要发展整合听觉和视觉信息的能力,是给予听觉经验或视觉经验就足够了呢?还是听-视组合经验更具有决定性作用?2.多感觉神经元接受两种或两种以上的感觉输入投射,但是不同感觉通道是否存在相互影响?如果有,它们是通过什么途径来实现的?又是如何相互作用的(相互抑制或者相互促进)?3.既然感觉经验对多感觉整合具有重要影响,那么感觉整合的基本方式是否可为感觉经验所改变?对这些神经系统基本问题的探索清晰地勾画出了本论文研究的目的和它的意义。全文共由三部分组成,紧紧围绕着多感觉系统经验依赖的可塑性展开。第一部分:多感觉经验驱动上丘多感觉神经元整合能力的发展在猫和猴子上的研究发现,上丘神经元的多感觉整合能力并非天生就有,需要经历一个发育成熟的过程。在黑暗环境中长大的猫上发现,上丘神经元缺乏多感觉整合能力,神经元对多感觉刺激的反应(multisensory response)并不比最佳单模态反应(unisensory response)强,表明感觉经验对多感觉整合能力的获得是必需的。为此,我们想知道在黑暗环境中饲养至成年的猫,是否保持着发展多感觉整合的能力?如果是,需要怎样的感觉经验?实验在5只黑暗环境中饲养至成年的猫上进行(年龄:7-11月)。首先通过开颅手术,暴露相应的脑区,同时给猫装上固定头的装置(head holder)。当猫恢复健康后开始实验,每周一次。在每一次实验中,将麻醉动物固定于立体定位装置,先给予3-4个小时感觉刺激训练,之后进行神经元反应的记录。记录停止后,待猫完全恢复后送回暗房。5只猫给予不同的感觉刺激:两只猫(Cat1和Cat2)接受多感觉经验暴露训练(即视觉刺激和听觉刺激同时出现在一个固定位置),每次实验给予1800次视听组合刺激,刺激间隔6秒;另外两只猫(Cat3和Cat4)也在相同的位置接受相同的视觉刺激和听觉刺激,但是听觉刺激和视觉刺激在时间上是分开的(单模态经验),刺激间隔4秒,每次实验分别给1800次听觉和1800次视觉刺激;另外一只猫(Cat5)接受在时间上间隔4秒但在空间上随机出现的听觉和视觉训练,给予的刺激数量与Cat3和Cat4相同。实验结果显示,多感觉经验可以促进上丘神经元多感觉整合能力的发展,而单模态经验则不能。从Cat1和Cat2上所记录到的多感觉神经元显示,它们的整合能力随着所感知多感觉经验的积累而逐渐增强。而从其它三只猫(Cat3,Cat4和Cat5)上记录到的多感觉神经元,它们的多感觉整合能力并没有太大的改变,仍然维持在最初的状态。尽管一部分神经元在经历了短期的多感觉刺激后(2-3实验天,3600-5400次刺激),整合能力迅速得到发展,但成熟的多感觉整合能力的获得,需要经历相当长的暴露过程(15个实验天,27000次的多感觉刺激)。此外实验结果显示,只有那些听觉空间感受野和视觉空间感受野均包含多感觉刺激训练位置的神经元才能发展多感觉整合能力,而对于那些一个感觉模态(V or A)或者两个感觉模态(V and A)的感受野均位于多感觉刺激训练位置之外的神经元,由于相当于只接受单模态刺激或者不能感受到刺激训练,多感觉整合能力也不能得到发展。一旦多感觉神经元在暴露位置发展了多感觉整合能力,作为一种信号处理的计算法则,这种整合能力往往可以扩展到感受野的其它区域。实验结果表明,多感觉神经元只有经历了多感觉经验才能获得成熟的信息整合能力。重复性的多感觉刺激不断地改变着神经系统中处理不同感觉模态信息的方式,不断地调制多感觉信息之间相互易化与相互抑制的平衡,使相互抑制减弱,相互易化变强,从而使神经系统能够整合不同感觉模态的刺激。毫无疑问这可能只是其中一方面机制,新突触的形成,新神经环路的构建,无疑也是非常重要的。这些研究成果将为治疗感知觉信息整合有障碍的病人,或者恢复脑损伤病人正常的感觉运动整合,提供了重要的神经生物学实验依据。第二部分:短暂的多感觉刺激暴露增加单模态刺激的有效性不同的感觉信息通过各自的感觉通道输入到同一个神经元,使得多感觉神经元能够整合来自于不同感觉模态的信息,产生一个非常强的多感觉反应。我们不知道这种多感觉整合效应对感觉通道的信息传递效率或者对单模态刺激的有效性有什么样的影响。在这一部分实验中,我们探索了这个极为重要而又一直被忽略的问题。在猫上丘多感觉层,单感觉神经元和多感觉经元共同存在着。然而许多单感觉神经元也许是潜在的多感觉神经元,因为它们对一个感觉模态刺激的反应,可以被另一个感觉模态刺激的同时存在所影响。把这些单感觉神经元暴露于多感觉刺激(几分钟),我们发现在将近1/3的单模态神经元中,原先沉默的感觉模态可以被激活,感觉通道的效率增强,导致同样的单模态刺激可以驱动神经元产生动作电位反应(暴露前不能)。例如,一个原先只对听觉刺激起反应的单模态神经元经过短暂的听-视联合暴露训练后,对视觉刺激也开始产生反应,尽管反应强度还比较弱。被诱导出的反应往往可以维持超过20min。令人意外的是,8个单模态神经元在经历了短暂地有效单模态刺激暴露训练后,也开始对来自于沉默模态的刺激产生反应(例如一个视觉神经元,在经历了短暂的视觉经验暴露后,对听觉刺激也起反应了)。但是多感觉暴露训练的效应显然明显优于有效单模态暴露训练,表现为视-听联合暴露训练所诱导产生的单模态反应更强,持续时间更长。多感觉刺激暴露可以把单感觉神经元诱导成多感觉神经元,那么对多感觉神经元的效应又是什么呢?48个多感觉神经元被用于此项研究,结果显示,多感觉训练也可以增强多感觉神经元对单模态刺激的敏感性(尤其是对弱感觉通道的感觉刺激),从而提升单模态刺激信息单独作用时的有效性。即多感觉刺激暴露后,单模态刺激的反应增强,放电频数增高,反应更加稳定。我们的研究结果显示了多感觉整合的又一个可能是非常重要规则。Hebbian的LTP(长时程增强效应)/STDP(spike timing dependent plasticity,放电反应时间依赖性可塑性)机制,可能在感觉通道效率的增强中起着非常重要的作用。考虑到多感觉刺激暴露后多感觉神经元普遍存在着自发放电的增强,提示神经元的兴奋性增强,这有可能是不同模态的感觉通道或信号相互作用的另外一种模式。第三部分:不同感觉模态联合刺激序列诱导多感觉整合特性变化多感觉神经元往往只能整合时间上一致或临近的多感觉刺激,如果两个不同模态的感觉刺激在时间上完全分开,神经元就不能对它们进行整合。我们想知道特殊的感觉训练是否可以改变这种状况?实验在正常环境中长大的三只猫上进行,当记录到一个多感觉神经元后,首先确定神经元的反应特性,确定两个序列刺激之间的时间间隔(SOA,stimulusonset asynchrony),使前后两个反应能完全分开,但又不分离太远,然后给予50-80次的这种序列的视-听刺激训练。我们发现神经元很快适应了这种带有事件性质的刺激组合。前后两个刺激所引起的反应逐渐靠近,开始融合,具体表现为前一个反应的时程(duration)延长,后一个反应的潜伏期(Latency)缩短,表现出神经元能够整合这种序列的多感觉刺激。这种效应只局限于序列的多感觉联合刺激,而序列的同一模态联合暴露训练则很少产生类似的效果。当暴露训练停止后,相应的单模态刺激反应再次被评价,发现效应仍然存在.神经元对第一个刺激的单模态反应产生了一种类似预测的效应,反应时程变长,放电变强,似乎在等待另一个感觉刺激的到来。但序列的多感觉联合暴露训练对第二个刺激的单模态反应没有太大的影响。实验结果表明,多感觉神经元能够快速地适应外在环境变化,从而能够在一个更加宽广的范围内对多模态信息进行整合。同时如果我们把序列出现的两个不同模态的刺激看作是一个微型的事件,那么多感觉神经元对这一微型多模态感觉事件的编码,也许代表着神经系统的普遍方式,这需要有更多的实验来验证。STDP(spike timing dependent plasticity,放电反应时间依赖性可塑性)理论模型至少从一个方面很好地解释了我们所得到的结果。根据这个理论模型,重复地暴露于短暂时间间隔的两个刺激,后面一个刺激的反应能够反馈性地支持前面一个刺激的反应,使前一个反应变强,放电时程变长。而多感觉神经元能够整合不同感觉模态信息这一标志性特性则是另外一个非常重要的原因。但是当先后的两个刺激均来自于相同的感觉模态时,由于它们通过相同(或部分相同)的感觉通道到达同一个神经元,感觉通路受到易化和抑制两种效应的共同作用,其次多感觉神经元无法整合来自同一模态的不同刺激(通常会发生抑制),使得序列的两个反应进行融合成为不可能。更多还原

【Abstract】 Animals receive different sensory information through different sensory systems. For example, the brain receives the visual and auditory information which is firstly encoded by the relative sensory apparatuses then conducted through the respective afferent pathway. However, the external events and stimuli are often composed of multiple sensory cues, which require the brain to be capable of synthesizing the different sensory inputs to reflect the nature of them. Much less is known about how the brain integrates the different modality sensory inputs. Cat SC provides the ideal place for such research, wherein multiple sensory inputs converging into the deep layer leads to the high proportion distribution of multisensory neurons. SC multisensory neurons play the pivotal role in the processes of the integration of the multiple sensory information and the sensory-motor integration, which is believed to increase the likelihood of detecting and identifying events or objects of interest.In the past more than 30 years, the multisensory integration in the nervous system was well examined and some fundamental integrative principles were extracted. Much attention was paid to the importance of information integration in the brain by more and more scientists, and many of them even joined this research field. Despite this, it must be said that we know little about multisensory integration in the neural system. Some questions are listed below.One: What kinds of sensory experience are of particular importance for the development of multisensory integration while the previous studies demonstrated acquirement of the multisensory integration capability was dependent on the sensory experience? An example is provided that if a naive auditory-visual multisensory neuron tries to develop its auditory-visual multisensory integration capability, what kind of sensory experience is needed, visual, auditory, unpaired auditory-visual or paired auditory-visual experience?Two: We know that multisensory neurons received more than one modality-specific sensory input. We did not know whether / how different sensory channels influence with each other? Three: Whether or not multisensory integrative principles can be altered by the particular sensory experience since that is so critical to the multisensory integration?The purpose of my research is to explore these fundamental and meaningful issues of the neural system. The research was conducted on the three different fields, which focused on the sensory - dependent plasticity of the multisensory system.Chapter 1: Initiating the Development of Multisensory Integration by Manipulating Sensory ExperienceThe studies in the cats and monkeys reported that multisensory integration was not an innate characteristic of SC neurons. Acquirement of multisensory integration is sensory dependent. The evidence from the dark reared cats showed that multisensory neurons failed to integrate cross-modal stimuli and response to the cross-modal stimuli was not significantly better than response to the more effective component stimulus. We wonder whether sensory experience could craft the multisensory neurons in dark reared cat SC to develop the ability to integrate the cross-modal stimuli? The experiment was conducted in five adult dark reared cats (postnatal months: 7-11). A recording chamber was stereotaxically placed over a craniotomy to provide access to the SC via the overlying cortex and attached to the skull.Weekly recording sessions began after a postsurgical recovery period of at least 7 days. After anesthesized and paralyzed, the animal was fixed in the stereotaxic apparatus. Different animals were exposed to different stimulus configurations. Two animals (Cat 1 and Cat 2) were exposed to spatiotemporally coincident visual and auditory stimuli (i.e., cross-modal stimuli) at a 6s interstimulus-interval (3hrs/day). Other two animals (Cat 3 and Cat 4) were given the exposure to these same visual stimulus and auditory stimulus for the same number of iterations in these same spatial locations, but with interleaved presentation of these modality-specific stimuli. The fifth animal (Cat 5) was also exposed to these modality-specific stimuli, but in spatial-temporal random way.The result demonstrated that multisensory experience rather than unisensory experience drove the developing of the multisensory integration. The data obtained from the Cat1 and Cat2 demonstrated that multisensory neurons’ integrative capability gradually enhanced with the accumulation of multisensory experience. Although a couple of multisensory neurons rapidly developed the multisensory integration capability after only experienced with the quite shorted-term period of cross-modal exposure (2-3 sessions: 3600-5400 trials), the acquirement of mature multisensory integration capability was found to need a large number of cross-modal stimuli exposure trials (more than 27000 exposure trials). The key experiential factor was repeated exposure to the relevant stimuli, and this required that the multiple receptive fields of a multisensory neuron encompass the cross-modal exposure site. Simple exposure to the individual components of a cross-modal stimulus was ineffective in this regard. Furthermore, once a neuron acquired multisensory integration capabilities at the exposure site, it generalized this experience to other locations, albeit with lowered effectiveness.The present data showed that multisensory integration can develop only after experience with multisensory experience. One of the speculations is that long-term repeated cross-modal stimuli shifted the balance between excitatory and inhibitory interaction of different sensory inputs, with the decrement of the mutual inhibition and the enhancement of the mutual facilitation. Also, there was high likelihood that acquirement of multisensory integration was the consequence of the formation of new synapses and the construction of new neural circuits. Our results provided the critical experimental evidence for curing the patients with incapability of multisensory integration and for recovering the multisensory integration capability of patients with brain injury.Chapter 2: Short-term Exposure to Cross-modal Stimuli Increased the Effectiveness of Modality-specific StimulusMultisensory neurons fuse the different sensory inputs to yield the robust multisensory response. What is the impact on the individual sensory channels afforded by this integrative product? Here we will examine this quite critical but always neglected issue. Multisensory neurons and unisensory neurons coexisted in cat SC. We believe that some proportion of unisensory neurons may be the potential multisensory neurons due to that the unisensory response was modulated by the presentation of another modality stimulus.Here we found that after the experience with short-term cross-modal exposure, nearly one third of unisensory neurons studied became multisensory neurons and the spike response was evoked by the stimulus in the initially silent sense. For example, the neuron initially viewed as the auditory neuron yielded the spike response to the visual stimulus after the short-term period of auditory-visual stimuli exposure. Newly evoked responses were characterized by the low magnitude. Most of them could last for more than 20 min. 8 unisensory neurons, unexpectedly, begin to respond to the stimulus in the silent modality after repeatedly exposed to the effective modality-specific stimulus. Despite that, it was obvious that cross-modal exposure has more effectiveness in activating "the silent modality" than the effective modality-specific exposure. The magnitude of evoked response from the new activated modality was better and the retaining time lasted longer following cross-modal exposure than following the modality-specific exposure. We also explored that the impact of cross-modal training on the multisensory neurons. This experiment was conducted in the 48 multisensory neurons. The result showed that cross-modal exposure would enhance the effectiveness of modality-specific stimulus (especially for initial weak effective stimulus) so that the same modality-specific stimulus could evoke the robust response after the exposure. The magnitude of the response was increased and the interstimulus interval was shortened and the response variability was decreased.Our research results revealed the important influence of multisensory integration on sensory channels that multisensory integration could enhance the efficacies of sensory channels. Hebbian LTP mechanism / STDP (Spike timing dependent plasticity) may underlie it. Considering the enhancement of spontaneous firing rate after cross-modal exposure, change in the neuronal excitatory state induced by the cross-modal exposure and modality-specific exposure, was also likely to engage in this process. Chapter 3: Adult plasticity in multisensory neurons: Short-term experience-dependent changes in the superior colliculusMultisensory neurons can integrate the cross-modal stimuli presented in space and time proximity. When the longer SOA (Stimulus onset asynchrony) was used, the multisensory neuron was incapable of synthesizing the cross-modal stimuli and the responses evoked by the different stimuli were separated temporally. We wonder whether this situation can be changed by the sensory experience.The experiment was conducted on the three cats. After a multisensory neuron was isolated, the response properties were examined and the SOA was determined. Repeated sequential cross-modal exposure training (50-80 trials) was given. It was found that neurons rapidly adapted its response to this stimuli configuration. This short-term plasticity was evident as a rapid increase in the magnitude and duration of responses to the first stimulus, and a shortening of the latency and increase in magnitude of the responses to the second stimulus when they are presented in sequence. The result was that the two responses appeared to merge. These changes were stable in the absence of experience with competing stimulus configurations, outlasted the exposure period, and could not be induced by equivalent experience with sequential within-modal (visual-visual or auditory-auditory) stimuli.The results showed that multisensory neurons in adult Cat SC can rapidly adapt information processing capabilities to deal with the environmental change so that multisensory integration can take place in the extensive range of the cross-modal stimuli configuration. In addition, if the sequential presentation of the cross-modal stimuli is treated as a miniature event, neurons seems to be able to encode the context, which may represent the way in which high cognitive level of neural system encode the similar event. STDP (Spike Timing Dependent Plasticity) theoretical model at least partly provides the explanation of the present results. According to this model, the second response could provide the preceding stimuli with the feedback supporting during training, which results in the enhancement of the response to the preceding stimulus. Multisensory neurons’ capability of integrating the cross-modal stimuli was likely to be another important support. However, if both components of the sequential stimuli are the same modality, the sensory channel afforded by the facilitation and inhibition due to the sharing the same or most of sensory channel by the preceding stimulus and the following stimulus can not effectively be potentiated which made it impossible to integrate the sequential within-modality stimuli.更多还原