Department of Physiology, The University of Tokyo School of Medicine, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
Eur J Neurosci. 2010 Aug;32(4):659-67. doi: 10.1111/j.1460-9568.2010.07320.x.
The perirhinal cortex, which is critical for long-term stimulus-stimulus associative memory, consists of two cytoarchitectonically distinct subdivisions: area 35 (A35) and area 36 (A36). Previous electrophysiological studies suggested that macaque A36 is involved in both association and retrieval processes during a visual pair-association task. However, the neuronal properties of macaque A35 have never been examined because A35 is located in a very narrow region, which makes it difficult to systematically record single-unit activity from there. In the present study, we overcame this technical difficulty for targeting A35 by combining magnetic resonance imaging-guided in-vivo localization with postmortem histological localization. This two-track approach enabled us to record from 181 A35 neurons in two macaque monkeys while they performed a pair-association task. Among these neurons, 64 showed stimulus-selective responses during the cue period (cue-selective neurons), whereas 18 did during the delay period (delay-selective neurons). As in A36, the responses of cue-selective neurons in A35 to paired associates were correlated. In both areas, these correlations were stronger in neurons showing delay selectivity than in those without delay selectivity. Notably, delay-selective neurons in A35 responded similarly to the optimal stimulus and its paired associate, whereas delay-selective neurons in A36 discriminated between them. However, these neurons in both areas discriminated the primary pair, consisting of the optimal stimulus and its paired associate, from other pairs, indicating that selectivity across pairs was maintained between the two areas. These results suggest that delay-selective neurons in A35 represent these paired stimuli as a single unitized item rather than two associated items.
研究人员发现，在一个视觉对偶联想任务中，猕猴的 A36 区参与了联想和提取过程。然而，由于 A35 区位于一个非常狭窄的区域，使得从该区域系统地记录单个神经元活动变得非常困难，因此，猕猴 A35 的神经元特性从未被研究过。在本研究中，我们通过结合磁共振成像引导的活体定位和死后组织学定位，克服了针对 A35 的这一技术难题。这种双轨方法使我们能够在两只猕猴执行对偶联想任务时，从 181 个 A35 神经元中进行记录。在这些神经元中，有 64 个在提示期（提示选择性神经元）表现出对刺激的选择性反应，而 18 个在延迟期（延迟选择性神经元）表现出对刺激的选择性反应。与 A36 区一样，A35 区中提示选择性神经元对配对联想的反应是相关的。在这两个区域中，在表现出延迟选择性的神经元中，这些相关性比没有延迟选择性的神经元更强。值得注意的是，A35 区的延迟选择性神经元对最佳刺激及其配对联想的反应相似，而 A36 区的延迟选择性神经元则能对它们进行区分。然而，这两个区域的这些神经元都能区分由最佳刺激及其配对联想组成的原始配对，以及其他配对，这表明两个区域之间的配对选择性得以保持。这些结果表明，A35 区的延迟选择性神经元将这些配对刺激作为一个单一的整体项目来表示，而不是两个相关的项目。
