Novel approaches to characterize neural responses to standard and non-standard sounds in humans: A single-trial MEG/EEG study of the auditory mismatch negativity.

一个牛仔在他的火堆旁休息，当一个接近的入侵者折断了一根小树枝时，他立即警觉起来。通过磁图或脑电图进行的非侵入性脑成像表明，这种在西部片中频繁出现的场景可能是有真实依据的。一种罕见的异常声音，这里是由树枝发出的，呈现在火发出的一系列频繁的标准声音中，从听觉皮层唤起所谓的失配负性（MMN），甚至在注意力集中前或睡眠期间，因此可能会提醒牛仔。在流行的解释中，MMN被认为反映了一个基于记忆的过程，在这个过程中，传入的感觉信息与由标准形成的听觉记忆的短暂痕迹进行比较。MMN的改变是各种神经退行性疾病和发育障碍的潜在指标。由于科学实验中MMN的信噪比通常很小，因此MMN是由对大量标准和偏差的响应的平均值推导而来。然而，牛仔应该在一次试验中更好地探测到异常的声音，否则他可能永远不会再享受火了。我们计划通过批判性地评估标准和偏差的作用来研究人类的MMN，不仅是在平均数据上，而且特别是在单次试验记录上。我们期望通过新的分析技术，即使用匹配跟踪算法对信号进行自适应逼近来达到这一目标。

A cowboy resting next to his fire is alerted immediately when an approaching intruder breaks a tiny twig. Non-invasive brain imaging by means of magneto- or electroencephalography suggests that this frequent scene in Westerns may have a real basis. An infrequent deviant sound, here produced by the twig, presented in a sequence of frequent standard sounds produced by the fire, evokes a so-called mismatch negativity (MMN) from the auditory cortex, even pre-attentively or during sleep, and thus could alert the cowboy. In the prevailing interpretation, the MMN is considered to reflect a memorybased process, in which the incoming sensory information is compared with a transient trace of the auditory memory formed by the standards. Alterations of the MMN are a potential indicator of various neurodegenerative diseases and developmental disorders. Since the signal-to-noise ratio of the MMN is typically small in scientific experiments, the MMN is derived from averages of responses to a large number of standards and deviants. The cowboy, however, should better detect the deviant sound on a single trial, as otherwise he might never enjoy a fire again. We plan to study the MMN in humans by critically assessing the roles attributed to the standards and the deviants not only on the average data, but, specifically, on the single-trial recordings. We expect to reach this aim by means of novel analysis techniques, the adaptive approximation of the signals using the matching pursuit algorithm.