Deficits in attention underpin many of the cognitive and neuropsychiatric features of Lewy body dementia. These attention-related symptoms remain difficult to treat and there are many gaps in our understanding of their neurobiology. An improved understanding of attention-related impairments can be achieved via mathematical modelling approaches, which identify cognitive parameters to provide an intermediate level between observed behavioural data and its underlying neural correlate. Here, we apply this approach to identify the role of impaired sensory evidence accumulation in the attention deficits that characterise Lewy body dementia. In 31 people with Lewy body dementia (including 13 Parkinson's disease dementia and 18 dementia with Lewy bodies cases), 16 people with Alzheimer's disease, and 23 healthy controls, we administered an attention task whilst they underwent functional 3T MRI. Using hierarchical Bayesian estimation of a drift diffusion model, we decomposed task performance into drift rate and decision boundary parameters. We tested the hypothesis that the drift rate - a measure of the quality of sensory evidence accumulation - is specifically impaired in Lewy body dementia, compared to Alzheimer's disease. We further explored whether trial-by-trial variations in the drift rate related to activity within the default and dorsal attention networks, to determine whether altered activity in these networks was associated with slowed drift rates in Lewy body dementia. Our results revealed slower drift rates in the Lewy body dementia compared to the Alzheimer's disease group, whereas the patient groups were equivalent for their decision boundaries. The patient groups were reduced relative to controls for both parameters. This highlights sensory evidence accumulation deficits as a key feature that distinguishes attention impairments in Lewy body dementia, consistent with impaired ability to efficiently process information from the environment to guide behaviour. We also found that the drift rate was strongly related to activity in the dorsal attention network across all three groups, whereas the Lewy body dementia group showed a divergent relationship relative to the Alzheimer's disease and control groups for the default network, consistent with altered default network modulation being associated with impaired evidence accumulation. Together, our findings reveal impaired sensory evidence accumulation as a specific marker of attention problems in Lewy body dementia, which may relate to large-scale network abnormalities. By identifying impairments in a specific sub-process of attention, these findings will inform future exploratory and intervention studies that aim to understand and treat attention-related symptoms that are a key feature of Lewy body dementia.

中文翻译：

---

路易体痴呆的感觉证据积累和网络功能受损

注意缺乏是路易氏体痴呆的许多认知和神经精神特征。这些与注意力有关的症状仍然难以治疗，并且我们对其神经生物学的理解还存在许多空白。可以通过数学建模方法来提高对注意力相关损伤的理解，该数学建模方法可以识别认知参数，从而在观察到的行为数据及其潜在的神经相关因素之间提供中间水平。在这里，我们应用这种方法来确定受损的感官证据积累在表征路易体痴呆症的注意力缺陷中的作用。在31位路易体痴呆患者中（包括13例帕金森氏病痴呆症和18例路易体痴呆患者），16例阿尔茨海默氏病患者和23位健康对照者，当他们进行功能性3T MRI时，我们执行了一项注意任务。使用漂移扩散模型的分层贝叶斯估计，我们将任务性能分解为漂移率和决策边界参数。我们测试了以下假设：与阿尔茨海默氏病相比，路易体痴呆患者的漂移率（一种衡量感觉证据积累质量的指标）特别受损。我们进一步探讨了漂移速率的逐项试验变化是否与默认和背侧注意网络内的活动相关，以确定这些网络中活动的改变是否与路易体痴呆的漂移速率降低相关。我们的结果显示，与阿尔茨海默氏病组相比，路易体痴呆症的漂移速度较慢，而患者组在决策范围上是等效的。相对于两个参数的对照，患者组都减少了。这突出了感官证据积累不足，这是区分路易氏体痴呆的注意障碍的关键特征，这与有效处理来自环境的信息以指导行为的能力受损相一致。我们还发现，在所有三个组中，漂移率均与背注意力网络中的活动密切相关，而路易体痴呆组与默认网络的阿尔茨海默氏病和对照组之间存在分歧关系，这与默认网络的变化一致调制与证据积累受损有关。一起，我们的发现表明，感觉证据积累受损是路易氏体痴呆中注意问题的特定标志，这可能与大规模网络异常有关。通过识别特定注意子过程中的障碍，这些发现将为将来的探索性研究和干预研究提供信息，这些研究旨在理解和治疗与注意力相关的症状，这是路易氏体痴呆症的关键特征。