Ruiz Tatiana, Baldwin Alex S, Spiegel Daniel P, Hess Robert, Farivar Reza
Research Institute of the McGill University Health Center, Montreal, QC, Canada.
Vision Sciences, Essilor R&D, Center for Innovation and Technology, Singapore, Singapore.
Front Neurol. 2019 Jul 18;10:767. doi: 10.3389/fneur.2019.00767. eCollection 2019.
The bulk of deficits accompanying mild traumatic brain injury (mTBI) is understood in terms of cortical integration-mnemonic, attentional, and cognitive disturbances are believed to involve integrative action across brain regions. Independent of integrative disturbances, mTBI may increase cortical noise, and this has not been previously considered. High-level integrative deficits are exceedingly difficult to measure and model, motivating us to utilize a tightly-controlled task within an established quantitative model to separately estimate internal noise and integration efficiency. First, we utilized a contour integration task modeled as a cortical-integration process involving multiple adjacent cortical columns in early visual areas. Second, we estimated internal noise and integration efficiency using the linear amplifier model (LAM). Fifty-seven mTBI patients and 24 normal controls performed a 4AFC task where they had to identify a valid contour amongst three invalid contours. Thresholds for contour amplitude were measured adaptively across three levels of added external orientation noise. Using the LAM, we found that mTBI increased internal noise without affecting integration efficiency. mTBI also caused hemifield bias differences, and efficiency was related to a change of visual habits. Using a controlled task reflecting cortical integration within the equivalent noise framework empowered us to detect increased computational noise that may be at the heart of mTBI deficits. Our approach is highly sensitive and translatable to rehabilitative efforts for the mTBI population, while also implicating a novel hypothesis of mTBI effects on basic visual processing-namely that cortical integration is maintained at the cost of increased internal noise.
轻度创伤性脑损伤（mTBI）伴随的大部分功能缺陷可从皮层整合的角度来理解——记忆、注意力和认知障碍被认为涉及大脑区域间的整合作用。独立于整合障碍之外，mTBI可能会增加皮层噪声，而这一点此前尚未被考虑到。高级别的整合缺陷极难测量和建模，这促使我们在一个既定的定量模型中利用一项严格控制的任务来分别估计内部噪声和整合效率。首先，我们采用了一项轮廓整合任务，该任务被建模为一个涉及早期视觉区域中多个相邻皮层柱的皮层整合过程。其次，我们使用线性放大器模型（LAM）来估计内部噪声和整合效率。57名mTBI患者和24名正常对照者执行了一项4AFC任务，他们必须在三个无效轮廓中识别出一个有效轮廓。在添加了三个水平的外部方向噪声的情况下，自适应地测量轮廓幅度的阈值。使用LAM，我们发现mTBI增加了内部噪声，而不影响整合效率。mTBI还导致了半视野偏差差异，并且效率与视觉习惯的改变有关。在等效噪声框架内使用一项反映皮层整合的受控任务使我们能够检测到可能是mTBI缺陷核心的计算噪声增加。我们的方法高度敏感，可转化为针对mTBI人群的康复努力，同时还暗示了一个关于mTBI对基本视觉处理影响的新假设——即皮层整合是以增加内部噪声为代价来维持的。
