The range of experimental questions that have been addressed using functional imaging methods has, until relatively recently, been constrained by the fact that it has not been possible to acquire fMRI responses to individually presented stimuli, and to separate those responses off-line according to variables such as stimulus and response type. This constraint is of particular relevance in studies of memory, where it is desirable to be able to separate individual trials according to the memory judgment made by the subject. Consequently, the development of event-related fMRI is an important technical progression, which should extend the range of questions that can be addressed using magnetic resonance imaging approaches. Two important examples of the application of this technique to memory research have recently been reported[1xBuilding memories: remembering and forgetting of verbal experiences as predicted by brain activity. Wagner, A.D. et al. Science. 1998; 281: 1188–1191Crossref | PubMedSee all References, 2xMaking memories: brain activity that predicts how well visual experience will be remembered. Brewer, J.B. et al. Science. 1998; 281: 1185–1187Crossref | PubMedSee all References]. In both of these studies, separate images were acquired while subjects observed individual stimuli. These images were then separated into two categories according to whether each stimulus was subsequently remembered or forgotten on a recognition-memory task that followed the acquisition phase. The logic of this approach is that any differences between these two newly formed categories may reflect neural activity that is related to successful memory encoding. The method itself is not new, and has been employed in a number of event-related potential studies of memory encoding. However, the application of this paradigm with functional magnetic resonance imaging has provided more detailed knowledge about the neural structures that are important for successful encoding into memory. For words, the areas predictive of successful retrieval were located in left-temporal and left prefrontal regions[1xBuilding memories: remembering and forgetting of verbal experiences as predicted by brain activity. Wagner, A.D. et al. Science. 1998; 281: 1188–1191Crossref | PubMedSee all References][1], while for pictures the equivalent areas were right prefrontal cortex as well as bilateral temporal regions[2xMaking memories: brain activity that predicts how well visual experience will be remembered. Brewer, J.B. et al. Science. 1998; 281: 1185–1187Crossref | PubMedSee all References][2]. These findings indicate that the neural basis of successful memory encoding varies according to stimulus type, as well as suggesting that there is at least partial overlap between the brain regions that are engaged during memory encoding and memory retrieval.

中文翻译：

---

事件相关 fMRI 和记忆

直到最近，使用功能成像方法解决的一系列实验问题一直受到以下事实的限制：无法获得对单独呈现的刺激的 fMRI 响应，并根据变量离线分离这些响应例如刺激和反应类型。这种约束在记忆研究中特别相关，在这种研究中，希望能够根据受试者做出的记忆判断来分离单个试验。因此，事件相关 fMRI 的发展是一项重要的技术进步，它将扩大使用磁共振成像方法可以解决的问题范围。最近报道了将此技术应用于记忆研究的两个重要例子[1xBuilding memory：记忆和遗忘由大脑活动预测的语言经验。瓦格纳，广告等。科学。1998年；281: 1188–1191Crossref | PubMed 查看所有参考文献，2x 记忆：预测视觉体验被记住的程度的大脑活动。布鲁尔，JB 等。科学。1998年；281: 1185–1187Crossref | PubMed 查看所有参考文献]。在这两项研究中，在受试者观察个体刺激的同时获取了单独的图像。然后根据每个刺激是随后在采集阶段之后的识别记忆任务中被记住还是遗忘，将这些图像分为两类。这种方法的逻辑是，这两个新形成的类别之间的任何差异都可能反映与成功的记忆编码相关的神经活动。方法本身并不新鲜，并已被用于许多与事件相关的记忆编码潜在研究中。然而，这种范式与功能磁共振成像的应用提供了关于神经结构的更详细的知识，这些神经结构对于成功编码到记忆中很重要。例如，预测成功检索的区域位于左颞叶和左前额叶区域[1x构建记忆：记忆和遗忘由大脑活动预测的语言经验。瓦格纳，广告等。科学。1998年；281: 1188–1191Crossref | PubMed查看所有参考文献][1]，而对于图片，等效区域是右前额叶皮层以及双侧颞区[2x制作记忆：预测视觉体验被记住的程度的大脑活动。布鲁尔，JB 等。科学。1998年；281：1185–1187Crossref | PubMed 查看所有参考文献][2]。这些发现表明，成功记忆编码的神经基础因刺激类型而异，并表明在记忆编码和记忆检索期间参与的大脑区域之间至少存在部分重叠。