Gallée Jeanne, Cordella Claire, Fedorenko Evelina, Hochberg Daisy, Touroutoglou Alexandra, Quimby Megan, Dickerson Bradford C
Program in Speech and Hearing Bioscience and Technology, Harvard University, Cambridge, MA 02138, USA.
Frontotemporal Disorders Unit, Department of Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02129, USA.
Brain Sci. 2021 Jan 20;11(2):130. doi: 10.3390/brainsci11020130.
"Functional communication" refers to an individual's ability to communicate effectively in his or her everyday environment, and thus is a paramount skill to monitor and target therapeutically in people with aphasia. However, traditional controlled-paradigm assessments commonly used in both research and clinical settings often fail to adequately capture this ability. In the current study, facets of functional communication were measured from picture-elicited speech samples from 70 individuals with mild primary progressive aphasia (PPA), including the three variants, and 31 age-matched controls. Building upon methods recently used by Berube et al. (2019), we measured the informativeness of speech by quantifying the content of each patient's description that was relevant to a picture relative to the total amount of speech they produced. Importantly, form-based errors, such as mispronunciations of words, unusual word choices, or grammatical mistakes are not penalized in this approach. We found that the relative informativeness, or efficiency, of speech was preserved in non-fluent variant PPA patients as compared with controls, whereas the logopenic and semantic variant PPA patients produced significantly less informative output. Furthermore, reduced informativeness in the semantic variant is attributable to a lower production of content units and a propensity for self-referential tangents, whereas for the logopenic variant, a lower production of content units and relatively "empty" speech and false starts contribute to this reduction. These findings demonstrate that functional communication impairment does not uniformly affect all the PPA variants and highlight the utility of naturalistic speech analysis for measuring the breakdown of functional communication in PPA.
“功能性交流”指个体在其日常环境中有效进行交流的能力，因此对于失语症患者而言，这是监测和进行治疗干预的一项至关重要的技能。然而，研究和临床环境中常用的传统对照范式评估往往无法充分捕捉这种能力。在本研究中，从70名患有轻度原发性进行性失语症（PPA）的个体（包括三种变体）以及31名年龄匹配的对照者的图片引发言语样本中测量了功能性交流的多个方面。基于贝鲁贝等人（2019年）最近使用的方法，我们通过量化每位患者描述中与图片相关的内容相对于他们所产生的言语总量，来测量言语的信息量。重要的是，在这种方法中，基于形式的错误，如单词发音错误、不寻常的词汇选择或语法错误，不会受到惩罚。我们发现，与对照组相比，非流利型变体PPA患者言语的相对信息量或效率得以保留，而语音变异型和语义变异型PPA患者产生的信息量明显较少。此外，语义变异型中信息量的减少归因于内容单元的产出较低以及自我指涉的偏离倾向，而对于语音变异型，内容单元产出较低以及相对“空洞”的言语和话语开头错误导致了这种减少。这些发现表明，功能性交流障碍并非对所有PPA变体均产生一致影响，并突出了自然主义言语分析在测量PPA中功能性交流障碍方面的效用。
