Department of Psychology, University of Cambridge, Cambridge, CB2 3EB, United Kingdom
Department of Psychology, University of Cambridge, Cambridge, CB2 3EB, United Kingdom.
J Neurosci. 2022 Jan 19;42(3):454-473. doi: 10.1523/JNEUROSCI.0182-21.2021. Epub 2021 Nov 23.
The ability to make decisions based on external information, prior knowledge, and evidence is a crucial aspect of cognition and may determine the success and survival of an organism. Despite extensive work on decision-making mechanisms/models, understanding the effects of alertness on neural and cognitive processes remain limited. Here we use EEG and behavioral modeling to characterize cognitive and neural dynamics of perceptual decision-making in awake/low alertness periods in humans (14 male, 18 female) and characterize the compensatory mechanisms as alertness decreases. Well-rested human participants, changing between full-wakefulness and low alertness, performed an auditory tone-localization task, and its behavioral dynamics were quantified with psychophysics, signal detection theory, and drift-diffusion modeling, revealing slower reaction times, inattention to the left side of space, and a lower rate of evidence accumulation in periods of low alertness. Unconstrained multivariate pattern analysis (decoding) showed a ∼280 ms delayed onset driven by low alertness of the neural signatures differentiating between left and right decision, with a spatial reconfiguration from centroparietal to lateral frontal regions 150-360 ms. To understand the neural compensatory mechanisms with decreasing alertness, we connected the evidence-accumulation behavioral parameter to the neural activity, showing in the early periods (125-325 ms) a shift in the associated patterns from right parietal regions in awake, to right frontoparietal during low alertness. This change in the neurobehavioral dynamics for central accumulation-related cognitive processes defines a clear reconfiguration of the brain networks' regions and dynamics needed for the implementation of decision-making, revealing mechanisms of resilience of cognition when challenged by decreased alertness. Most living organisms make multiple daily decisions, and these require a degree of evidence from both the environment and the internal milieu. Such decisions are usually studied under sequential sampling models and involve making a behavioral choice based on sensory encoding, central accumulation, and motor implementation processes. Since there is little research on how decreasing alertness affects such cognitive processes, this study has looked at the cognitive and neural dynamics of perceptual decision-making in people while fully awake and in drowsy periods. Using computational modeling of behavior and neural dynamics on human participants performing an auditory tone-localization task, we reveal how low alertness modulates evidence accumulation-related processes and its corresponding compensatory neural signatures.
基于外部信息、先验知识和证据做出决策的能力是认知的一个关键方面，可能决定生物体的成功和生存。尽管已经进行了广泛的决策机制/模型研究，但对警觉状态对神经和认知过程的影响仍知之甚少。在这里，我们使用 EEG 和行为建模来描述人类清醒/警觉度降低期间（14 名男性，18 名女性）感知决策的认知和神经动力学，并描述警觉度降低时的补偿机制。休息良好的人类参与者在完全清醒和警觉度降低之间切换，执行听觉音调定位任务，其行为动态通过心理物理学、信号检测理论和漂移扩散模型进行量化，结果显示警觉度降低时反应时间变慢、对空间左侧的注意力不集中以及证据积累速度降低。无约束的多变量模式分析（解码）显示，警觉度降低时，左/右决策的神经特征区分的潜伏期约为 280ms，神经特征从中央顶叶区域到外侧额区的空间重新配置在 150-360ms 之间。为了理解警觉度降低时的神经补偿机制，我们将证据积累行为参数与神经活动连接起来，结果表明，在早期（125-325ms），与清醒时右顶叶相关的关联模式转变为警觉度降低时的右额顶叶模式。这种与中央积累相关的认知过程的神经行为动力学的变化定义了大脑网络区域和动态的明确重新配置，这些区域和动态是实施决策所必需的，揭示了认知在受到警觉度降低挑战时的弹性机制。大多数生物体每天都会做出多次决策，这些决策需要来自环境和内部环境的一定程度的证据。这些决策通常在序列抽样模型下进行研究，涉及基于感觉编码、中央积累和运动实施过程做出行为选择。由于很少有研究关注警觉度降低如何影响这些认知过程，因此本研究研究了人类在完全清醒和困倦期间进行听觉音调定位任务时的感知决策的认知和神经动力学。我们使用人类参与者执行听觉音调定位任务的行为和神经动力学的计算模型，揭示了警觉度降低如何调节与证据积累相关的过程及其相应的补偿神经特征。
