Gold Joshua I
University of Pennsylvania Department of Neuroscience, Philadelphia, PA 19104, USA.
Novartis Found Symp. 2006;270:92-101; discussion 101-13.
Experience and perception are deeply intertwined. Experience, particularly early in life, shapes how sensory information is represented in the brain. Experience also establishes associations and can affect how sensory information guides behaviour. Central to these kinds of perceptual abilities are neural mechanisms that interpret, or decode, the brain's sensory representation, but little is known about how these decoding mechanisms depend on experience. Here I discuss several critical roles that experience might play in shaping these mechanisms. First, experience is likely to drive changes in neural connectivity to select the spatially and temporally distributed sensory signals that provide relevant information about a stimulus. Second, even the most relevant sensory signals provide incomplete information about the presence of a stimulus; also necessary is knowledge of the a priori probability of the stimulus, which must be learned from experience. Third, decoding noisy information is necessarily imperfect and therefore involves trade-offs like speed versus accuracy and false alarms versus misses. Experience is likely to provide ongoing feedback about the value of these trade-offs so that they might be adjusted appropriately. Each of these mechanisms appear to be capable of causing dramatic changes in sensitivity, response bias, response times and other manifestations of perceptual ability.
经验与感知紧密相连。经验，尤其是在生命早期，塑造了感官信息在大脑中的呈现方式。经验还建立起关联，并能影响感官信息如何引导行为。这些感知能力的核心是解释或解码大脑感官表征的神经机制，但对于这些解码机制如何依赖经验却知之甚少。在此，我将探讨经验在塑造这些机制中可能发挥的几个关键作用。首先，经验可能会驱动神经连接的变化，以选择那些在空间和时间上分布的感官信号，这些信号提供了有关刺激的相关信息。其次，即使是最相关的感官信号也只能提供关于刺激存在的不完整信息；刺激的先验概率知识同样必不可少，而这必须从经验中习得。第三，解码有噪声的信息必然存在不完美之处，因此涉及速度与准确性、误报与漏报等权衡。经验可能会持续提供关于这些权衡价值的反馈，以便对其进行调整。这些机制中的每一种似乎都能够在敏感度、反应偏差、反应时间以及感知能力的其他表现方面引发显著变化。
