The growing prevalence of neurodegenerative disorders associated with aging and cognitive decline has generated increasing cross-disciplinary interest in non-pharmacological interventions, such as computerized cognitive training (CCT), which may prevent or slow cognitive decline. However, inconsistent findings across meta-analytic reviews in the field suggest a lack of cross-disciplinary consensus and on-going debate regarding the benefits of CCT. We posit that a contributing factor is the lack of a theoretically-based taxonomy of constructs and representative tasks typically used. An integration of the Cattell-Horn-Carroll (CHC) taxonomy of broad and narrow cognitive factors and the Miyake unity-diversity theory of executive functions (EF) is proposed (CHC-M) as an attempt to clarify this issue through representing and integrating the disciplines contributing to CCT research. The present study assessed the utility of this taxonomy by reanalyzing the Lampit et al. (2014) meta-analysis of CCT in healthy older adults using the CHC-M framework. Results suggest that: 1) substantively different statistical effects are observed when CHC-M is applied to the Lampit et al. meta-analytic review, leading to importantly different interpretations of the data; 2) typically-used classification practices conflate Executive Function (EF) tasks with fluid reasoning (Gf) and retrieval fluency (Gr), and Attention with sensory perception; and 3) there is theoretical and practical advantage in differentiating attention and working-memory tasks into the narrow shifting, inhibition, and updating EF domains. Implications for clinical practice, particularly for our understanding of EF are discussed.

中文翻译：

---

元分析的计算机认知训练对执行功能的影响：分类结果认知因素的跨学科分类法。

与衰老和认知能力下降相关的神经退行性疾病的患病率上升，已经引起了人们对非药物干预措施（例如计算机认知训练（CCT））的跨学科兴趣的增长，这种干预可以预防或减缓认知能力下降。但是，该领域的荟萃分析综述中不一致的发现表明，缺乏跨学科共识，并且就CCT的益处仍存在争议。我们假定一个促成因素是缺乏基于理论的结构和通常使用的代表性任务的分类法。广义和狭义认知因素的Cattell-Horn-Carroll（CHC）分类法与执行功能的Miyake统一多样性理论（EF提议（CHC-M）的目的是通过代表和整合有助于CCT研究的学科来阐明这一问题。本研究通过重新分析Lampit等人，评估了该分类法的效用。（2014年）使用CHC-M框架对健康老年人进行CCT的荟萃分析。结果表明：1）将CHC-M应用于Lampit等人时，观察到实质上不同的统计效果。元分析审查，导致对数据的重要不同解释；2）通常使用的分类实践将执行功能（EF）任务与流体推理（Gf）和检索流利度（Gr）融合在一起，将注意力与感官知觉融合在一起; （3）将注意力和工作记忆任务区分为狭窄的转移，抑制和更新EF域具有理论和实践上的优势。讨论了对临床实践的影响，特别是对我们对EF的理解。