Department of Communication Sciences and Disorders, MGH Institute of Health Professions, Boston, MA.
Research Department, Moss Rehabilitation Research Institute, Elkins Park, PA.
J Speech Lang Hear Res. 2023 Feb 13;66(2):668-687. doi: 10.1044/2022_JSLHR-22-00251. Epub 2023 Feb 2.
Increasingly, mechanisms of learning are being considered during aphasia rehabilitation. Well-characterized learning mechanisms can inform "how" interventions should be administered to maximize the acquisition and retention of treatment gains. This systematic scoping review mapped hypothesized mechanisms of action (MoAs) and treatment ingredients in three learning-based approaches targeting naming in aphasia: errorless learning (ELess), errorful learning (EFul), and retrieval practice (RP). The rehabilitation treatment specification system was leveraged to describe available literature and identify knowledge gaps within a unified framework.
PubMed and CINHAL were searched for studies that compared ELess, EFul, and/or RP for naming in aphasia. Independent reviewers extracted data on proposed MoAs, treatment ingredients, and outcomes.
Twelve studies compared ELess and EFul, six studies compared ELess and RP, and one study compared RP and EFul. Hebbian learning, gated Hebbian learning, effortful retrieval, and models of incremental learning via lexical access were proposed as MoAs. To maximize treatment outcomes within theorized MoAs, researchers manipulated study ingredients including cues, scheduling, and feedback. Outcomes in comparative effectiveness studies were examined to identify ingredients that may influence learning. Individual-level variables, such as cognitive and linguistic abilities, may affect treatment response; however, findings were inconsistent across studies.
Significant knowledge gaps were identified and include (a) which MoAs operate during ELess, EFul, and RP; (b) which ingredients are active and engage specific MoAs; and (c) how individual-level variables may drive treatment administration. Theory-driven research can support or refute MoAs and active ingredients enabling clinicians to modify treatments within theoretical frameworks.
在失语症康复中,越来越多地考虑学习机制。特征明确的学习机制可以为“如何”进行干预提供信息,以最大限度地获得和保留治疗效果。本系统范围综述绘制了三种基于学习的命名失语症治疗方法(无错误学习(ELess)、错误学习(EFul)和检索练习(RP))中假设的作用机制(MoA)和治疗成分。利用康复治疗规范系统来描述现有文献,并在统一框架内确定知识空白。
在 PubMed 和 CINHAL 上搜索比较命名性失语症的 ELess、EFul 和/或 RP 的研究。独立评审员提取关于拟议的 MoA、治疗成分和结果的数据。
12 项研究比较了 ELess 和 EFul,6 项研究比较了 ELess 和 RP,1 项研究比较了 RP 和 EFul。Hebbian 学习、门控 Hebbian 学习、费力检索和通过词汇访问进行增量学习的模型被提出作为 MoA。为了在理论化的 MoA 内最大化治疗效果,研究人员操纵了研究成分,包括线索、调度和反馈。在比较效果研究中检查了结果,以确定可能影响学习的成分。个体水平变量,如认知和语言能力,可能会影响治疗反应;然而,研究结果不一致。
确定了重大的知识空白,包括:(a)ELess、EFul 和 RP 期间哪些 MoA 起作用;(b)哪些成分是有效的,并且参与特定的 MoA;以及(c)个体水平变量如何驱动治疗管理。基于理论的研究可以支持或反驳 MoA 和有效成分,使临床医生能够在理论框架内修改治疗方法。