Cognition and Brain Plasticity Group, Bellvitge Biomedical Research Institute - IDIBELL, L'Hospitalet de Llobregat, Barcelona, Spain.
Department of Psychology, University of Alabama at Birmingham, Birmingham, AL, USA; Moss Rehabilitation Research Institute, Elkins Park, PA, USA.
Cortex. 2017 Aug;93:12-27. doi: 10.1016/j.cortex.2017.04.020. Epub 2017 May 4.
Human learners can resolve referential ambiguity and discover the relationships between words and meanings through a cross-situational learning (CSL) strategy. Some people with aphasia (PWA) can learn word-referent pairings under referential uncertainty supported by online feedback. However, it remains unknown whether PWA can learn new words cross-situationally and if such learning ability is supported by statistical learning (SL) mechanisms. The present study examined whether PWA can learn novel word-referent mappings in a CSL task without feedback. We also studied whether CSL is related to SL in PWA and neurologically healthy individuals. We further examined whether aphasia severity, phonological processing and verbal short-term memory (STM) predict CSL in aphasia, and also whether individual differences in verbal STM modulate CSL in healthy older adults. Sixteen people with chronic aphasia underwent a CSL task that involved exposure to a series of individually ambiguous learning trials and a SL task that taps speech segmentation. Their learning ability was compared to 18 older controls and 39 young adults recruited for task validation. CSL in the aphasia group was below the older controls and young adults and took place at a slower rate. Importantly, we found a strong association between SL and CSL performance in all three groups. CSL was modulated by aphasia severity in the aphasia group, and by verbal STM capacity in the older controls. Our findings indicate that some PWA can preserve the ability to learn new word-referent associations cross-situationally. We suggest that both PWA and neurologically intact individuals may rely on SL mechanisms to achieve CSL and that verbal STM also influences CSL. These findings contribute to the ongoing debate on the cognitive mechanisms underlying this learning ability.
人类学习者可以通过跨情境学习（CSL）策略来解决指代歧义并发现单词和词义之间的关系。一些失语症患者（PWA）在在线反馈支持下，可以在不确定的情况下学习单词-指代对。然而，目前尚不清楚 PWA 是否可以跨情境学习新单词，以及这种学习能力是否受到统计学习（SL）机制的支持。本研究考察了 PWA 是否可以在没有反馈的情况下在 CSL 任务中学习新的单词-指代映射。我们还研究了 CSL 在 PWA 和神经健康个体中是否与 SL 相关。我们进一步研究了在失语症中，失语症的严重程度、语音处理和言语短期记忆（STM）是否可以预测 CSL，以及言语 STM 的个体差异是否可以调节健康老年人的 CSL。16 名慢性失语症患者接受了 CSL 任务，该任务涉及暴露于一系列单独的歧义学习试验和语音分割的 SL 任务。将他们的学习能力与 18 名老年对照组和 39 名年轻成年人进行比较，这些成年人是为任务验证而招募的。失语症组的 CSL 低于老年对照组和年轻成年人，且速度较慢。重要的是，我们发现所有三组之间的 SL 和 CSL 表现之间存在很强的关联。在失语症组中，CSL 受到失语症严重程度的调节，而在老年对照组中，CSL 受到言语 STM 能力的调节。我们的发现表明，一些 PWA 可以保留跨情境学习新单词-指代关联的能力。我们认为，PWA 和神经健全的个体可能都依赖于 SL 机制来实现 CSL，言语 STM 也会影响 CSL。这些发现为这一学习能力的认知机制的持续争论做出了贡献。
