小柯机器人

大脑-表型模型不能用于违反样本刻板印象的个体
2022-08-28 22:34

近日,美国耶鲁大学R. Todd Constable、Abigail S. Greene等研究人员合作发现,大脑-表型模型不能用于违反样本刻板印象的个体。该研究于2022年8月24日在线发表于国际一流学术期刊《自然》。

研究人员使用预测模型将大脑活动与表型联系起来,并在独立的数据上进行训练和测试,从而确保可推广性,并检查模型的失败。研究人员将这种数据驱动的方法应用于新的、临床和人口统计学上异质的数据集中的一系列神经认知措施,其结果在两个独立的、公开的数据集中得到了重复。在所有三个数据集中,研究人员发现模型反映的不是单一的认知结构,而是与社会人口学和临床协变量交织在一起的神经认知分数;也就是说,模型反映的是刻板印象,当应用于违背刻板印象的个体时,就会失败。模型的失败是可靠的,表型是特定的,并且可以在不同的数据集中推广。

总之,这些结果突出了"一刀切"建模方法的缺陷,以及有偏见的表型测量对所产生的大脑表型模型的解释和效用的影响。研究人员提出了一个解决这些问题的框架,以便这类模型可以揭示支撑特定表型的神经回路,并最终确定临床干预的个体化神经靶标。

据了解,大脑功能组织的个体差异跟踪一系列的特征、症状和行为。到目前为止,建立大脑与表型的线性关系模型的工作是假设一个单一的这种关系在所有的个体中都有普遍性,但模型并不是在所有的参与者中都同样有效。更好地了解模型在谁身上失效以及为什么失效,对揭示稳健、有用和无偏见的大脑表型关系至关重要。

附:英文原文

Title: Brain–phenotype models fail for individuals who defy sample stereotypes

Author: Greene, Abigail S., Shen, Xilin, Noble, Stephanie, Horien, Corey, Hahn, C. Alice, Arora, Jagriti, Tokoglu, Fuyuze, Spann, Marisa N., Carrin, Carmen I., Barron, Daniel S., Sanacora, Gerard, Srihari, Vinod H., Woods, Scott W., Scheinost, Dustin, Constable, R. Todd

Issue&Volume: 2022-08-24

Abstract: Individual differences in brain functional organization track a range of traits, symptoms and behaviours1,2,3,4,5,6,7,8,9,10,11,12. So far, work modelling linear brain–phenotype relationships has assumed that a single such relationship generalizes across all individuals, but models do not work equally well in all participants13,14. A better understanding of in whom models fail and why is crucial to revealing robust, useful and unbiased brain–phenotype relationships. To this end, here we related brain activity to phenotype using predictive models—trained and tested on independent data to ensure generalizability15—and examined model failure. We applied this data-driven approach to a range of neurocognitive measures in a new, clinically and demographically heterogeneous dataset, with the results replicated in two independent, publicly available datasets16,17. Across all three datasets, we find that models reflect not unitary cognitive constructs, but rather neurocognitive scores intertwined with sociodemographic and clinical covariates; that is, models reflect stereotypical profiles, and fail when applied to individuals who defy them. Model failure is reliable, phenotype specific and generalizable across datasets. Together, these results highlight the pitfalls of a one-size-fits-all modelling approach and the effect of biased phenotypic measures18,19,20 on the interpretation and utility of resulting brain–phenotype models. We present a framework to address these issues so that such models may reveal the neural circuits that underlie specific phenotypes and ultimately identify individualized neural targets for clinical intervention.

DOI: 10.1038/s41586-022-05118-w

Source: https://www.nature.com/articles/s41586-022-05118-w

Nature:《自然》,创刊于1869年。隶属于施普林格·自然出版集团,最新IF:69.504
官方网址:http://www.nature.com/
投稿链接:http://www.nature.com/authors/submit_manuscript.html


本期文章:《自然》:Online/在线发表

分享到:

0