Department of Neurobiology and Anatomy, Wake Forest University School of Medicine Winston Salem, NC, USA.
Front Psychol. 2011 Jul 4;2:144. doi: 10.3389/fpsyg.2011.00144. eCollection 2011.
The superior colliculus (SC) integrates information from multiple sensory modalities to facilitate the detection and localization of salient events. The efficacy of "multisensory integration" is traditionally measured by comparing the magnitude of the response elicited by a cross-modal stimulus to the responses elicited by its modality-specific component stimuli, and because there is an element of randomness in the system, these calculations are made using response values averaged over multiple stimulus presentations in an experiment. Recent evidence suggests that multisensory integration in the SC is highly plastic and these neurons adapt to specific anomalous stimulus configurations. This raises the question whether such adaptation occurs during an experiment with traditional stimulus configurations; that is, whether the state of the neuron and its integrative principles are the same at the beginning and end of the experiment, or whether they are altered as a consequence of exposure to the testing stimuli even when they are pseudo-randomly interleaved. We find that unisensory and multisensory responses do change during an experiment, and that these changes are predictable. Responses that are initially weak tend to potentiate, responses that are initially strong tend to habituate, and the efficacy of multisensory integration waxes or wanes accordingly during the experiment as predicted by the "principle of inverse effectiveness." These changes are presumed to reflect two competing mechanisms in the SC: potentiation reflects increases in the expectation that a stimulus will occur at a given location relative to others, and habituation reflects decreases in stimulus novelty. These findings indicate plasticity in multisensory integration that allows animals to adapt to rapidly changing environmental events while suggesting important caveats in the interpretation of experimental data: the neuron studied at the beginning of an experiment is not the same at the end of it.
上丘（SC）整合来自多个感觉模态的信息，以促进显著事件的检测和定位。“多感觉整合”的有效性传统上是通过比较跨感觉刺激引起的响应与其模态特异性成分刺激引起的响应的大小来衡量的，由于系统中存在一定的随机性，这些计算是使用实验中多次刺激呈现的响应值平均值进行的。最近的证据表明，SC 中的多感觉整合具有高度的可塑性，这些神经元适应特定的异常刺激模式。这就提出了一个问题，即在具有传统刺激模式的实验中是否会发生这种适应；也就是说，在实验开始和结束时，神经元的状态及其整合原则是否相同，或者即使测试刺激是伪随机交错的，它们是否会因为暴露于测试刺激而发生变化。我们发现，在实验过程中，单感觉和多感觉反应确实会发生变化，而且这些变化是可以预测的。最初较弱的反应倾向于增强，最初较强的反应倾向于习惯化，并且多感觉整合的效果在实验过程中相应地增减，正如“反效原则”所预测的那样。这些变化被认为反映了 SC 中的两种竞争机制：增强反映了对给定位置出现刺激的预期相对于其他位置的增加，而习惯化反映了刺激新奇性的降低。这些发现表明多感觉整合具有可塑性，使动物能够适应快速变化的环境事件，同时对实验数据的解释提出了重要的警告：在实验开始时研究的神经元在实验结束时并不相同。
