一般口演（Oral）

Oral Learning, Memory and Plasticity-2 一般口演学習・記憶・可塑性-1

7月27日（土）14:20～14:35　第9会場（朱鷺メッセ　3F　306+307） 3O-09a1-1 Neurophysiology of hippocampus-independent memories in the macaque posterior parietal cortex Lei Wang（Wang Lei）¹，Shuzhen Zuo（Zuo Shuzhen）¹，Kofi Appiah（Appiah Kofi）²，Yudian Cai（Cai Yudian）¹，Zhiyong Jin（Jin Zhiyong）¹，Makoto Kusunoki（Kusunoki Makoto）^3,4，Yong-di Zhou（Zhou Yong-di）^1,6，Sze Chai Kwok（Kwok Sze Chai）^1,5,6 ¹Shanghai Key Laboratory of Brain Functional Genomics, Key Laboratory of Brain Functional Genomics Ministry of Education, School of Psychology and Cognitive Science, East China Normal University, China ²Department of Computing, Sheffield Hallam University, UK ³MRC Cognition and Brain Sciences Unit, Cambridge, UK ⁴Department of Experimental Psychology, University of Oxford, UK ⁵Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, China ⁶NYU-ECNU Institute of Brain and Cognitive Science at NYU Shanghai, China Recent fMRI studies on humans have revealed that rapid new memories formation is associated with BOLD activation and structural changes in the posterior parietal cortex, suggesting a hippocampus-independent neocortical memory engram (Brodt et al., PNAS, 2016; Brodt et al., Science, 2018). These cortices are also engaged as part of a social processing network that shows specific neural responses to viewing conspecifics in the macaques (Sliwa et al., Science, 2017). We then asked whether such experience-induced memories might contain content or behavioral-relevance specificity and how they might evolve over repetitive exposures (i.e., repetition suppression of complex experiences) during natural viewing. To address these questions, we recorded multi-unit neuronal activities using an array of 32-channal chronically implanted, independently movable micro-electrodes (SC32, Gray Matter Research) on two macaque monkeys while they underwent a natural viewing experiment. In each recording session, we had the macaques view three different videos each for 30 repetitions. We acquired neuronal data from 60 repetitions for each unique video (acquired across two consecutive days). In total, we used 18 unique videos and acquired data from 1080 video-viewing trials. The videos were all 30-s long and classified as either containing depiction of other primate animals, or depiction of non-primate animals, or of plain scenery (Category: primates/non-primates/scenery). We also manipulated contextual changes among the videos (Context: no boundary/one-boundary/two-boundary). The monkeys' eye movements during natural viewing were recorded at 240Hz by an iScan infrared eye-tracking system. In combination of convolutional neural network (CNN) models trained for object and feature classification, we used the oculomotor and spike data to classify parietal neurons by their selectivity to classes of visual contents and to verify the repetition suppression phenomenon during memory formation. These results help elucidate the neurophysiological basis of the recently discovered hippocampus-independent neocortical memory engram in the primates.		7月27日（土）14:35～14:50　第9会場（朱鷺メッセ　3F　306+307） 3O-09a1-2 Mnemonic rigidity revealed in macaques: A two-step mechanism underlying temporal-order judgement of naturalistic cinematic events Shuzhen Zuo（Zuo Shuzhen）¹，Lei Wang（Wang Lei）¹，Junghan Shin（Shin Junghan）²，Yudian Cai（Cai Yudian）¹，Zhiyong Jin（Jin Zhiyong）¹，Sangwan Lee（Lee Sangwan）³，Makoto Kusunoki（Kusunoki Makoto）^4,5，Yongdi Zhou（Zhou Yongdi）^1,7，SzeChai Kwok（Kwok SzeChai）^1,6,7 ¹Shanghai Key Laboratory of Brain Functional Genomics, Key Laboratory of Brain Functional Genomics Ministry of Education, School of Psychology and Cognitive Science, East China Normal University, Shanghai, China ²Program of Brain and Cognitive Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea ³Department of Bio and Brain Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea ⁴MRC Cognition and Brain Sciences Unit, Cambridge, UK ⁵Department of Experimental Psychology, University of Oxford, Oxford, UK ⁶Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China ⁷NYU-ECNU Institute of Brain and Cognitive Science at NYU Shanghai, Shanghai, China One important aspect of episodic memory is to remember the order in which events occurred. Previous studies have demonstrated that nonhuman primates possess the ability in making temporal order memory judgement. The underlying mechanism for retrieving episodic temporal information in macaque monkeys is however unclear. To address this question, we trained six macaque monkeys with a temporal order judgement (TOJ) task using naturalistic videos. In Experiment 1, monkeys performed 5000 TOJ trials with trial-unique videos. In each trial, they watched a video of about 10-s comprising two across-context clips and, after a 2-s retention delay, performed a temporal order judgement between two frames extracted from the video. We examined the effect of context shift by modelling the reaction time distributions in terms of threshold and information processing speed using the LATER (Linear Approach to Threshold with Ergodic Rate) model. The LATER results showed that context shifts accelerate the rate of accumulating information rather than by altering the decision-threshold for memory judgement, indicating their ability in detecting contextual change points in the stream of cinematic material. Moreover, regression analyses showed that monkeys responded significantly faster to frames that are extracted from earlier segments than those located at latter segments of the videos, suggesting that the monkeys may serially replay the videos in order to identify the target frame. This pattern of results was replicated with a new set of 8-s videos in Experiment 2, in which we systematically controlled for temporal similarity between conditions. Multi-unit activities are simultaneously recorded with semi-chronically implanted 32 independently movable microelectrodes (SC32, Gray Matter Research) in the medial posterior parietal cortex on two of the monkeys. We will present multi-unit activity and local field potential data in relation to how monkeys can segment episodic events in a coarse timescale like humans, as well as how they resort to inefficient serial-replay mechanisms when prominent temporal-contextual markers are absent during temporal-order retrieval.

7月27日（土）14:50～15:05　第9会場（朱鷺メッセ　3F　306+307） 3O-09a1-3 Differential contributions of subareas in the macaque medial temporal lobe to the context-dependent use of item-location association memory Cen Yang（Yang Cen）^1,4，Yuji Naya（Naya Yuji）^1,2,3,5 ¹Center for Life Sciences, Peking University, Beijing, China ²School of Psychological and Cognitive Sciences, Peking University, Beijing, China ³IDG/McGovern Institute for Brain Research, Peking University, Beijing, China ⁴Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China ⁵Interdisciplinary Institute of Neuroscience and Technology, Zhejiang University, Hangzhou, China The declarative memory system equips us to remember the information of past knowledge so as to use it for future action. Importantly, while numerous studies have investigated its encoding and retrieval processes, there is still room for exploration as to how retrieved information is used. On the last JNNS meeting, we reported neurophysiological evidence suggesting involvements of macaque medial temporal lobe (MTL) in the retrieval of item-location memory and its usage using a newly-devised memory task. However, specific functions of MTL subareas are still unclear. In the present study, we compared neuronal activities among the three MTL areas and area TE (HPC, n = 456; PRC, n = 319; PHC, n = 232; TE, n = 141) using the same memory task. In the task, 2 sets of 4 visual items were used as I-Cue stimuli. Each I-Cue stimulus was associated with one particular location out of four relative to an O-Cue stimulus. In each trial, an I-Cue and an O-Cue were sequentially presented with a delay in-between. The monkeys were required to saccade to the target location in accordance with a combination of the two cues. Substantial number of cells exhibited significant (P < 0.01, one-way ANOVA) stimulus-selective activities to I-Cue stimuli in all areas (I-Cue cells, 10-33% of the recorded cells). The I-Cue cells showed significantly correlated responses to the I-Cue stimuli that were assigned to the same locations in MTL areas (P < 0.001, Wilcoxon's signed-rank test), but not in TE (P = 0.11). Time courses of the population-averaged correlation coefficients showed that the association signal appeared first in PRC before HPC and PHC. Next we examined the neuronal activities related with target locations after O-Cue presentation using three-way ANOVA. We found that a proportion of the target-location cells (P < 0.01) was much larger in HPC (19%) compared with other areas (4-7%). To examine a relationship between the activities of HPC target-location cells (n = 47) and monkeys' behaviors, we compared neuronal activities in error trials with those in correct trials, and found that the activities in error trials were more correlated to the positions that the monkeys chose (r = 0.26, partial correlation coefficient) rather than to the positions expected by a combination of I-Cue and O-Cue (r = 0.12) (P < 0.0019, Wilcoxon's signed-rank test). These results may suggest that PRC contribute to the retrieval, while HPC contribute to its usage for the following action.		7月27日（土）15:05～15:20　第9会場（朱鷺メッセ　3F　306+307） 3O-09a1-4 Activation of Nucleus Reuniens is necessary for acquisition, but not consolidation and retrieval of trace fear conditioning. Yu Ju Lin（Lin Yu Ju） Institute of Systems Neuroscience, National Tsing Hua University, Hsinchu, TAIWAN. In Pavlovian fear conditioning, a conditioned stimulus (CS), such as a tone, is associated with an unconditioned stimulus (US), such as a footshock. Fear conditioning is robust in that animals can learn the CS-US association after a few pairings. The temporal relationship between the CS and the US is crucial. In delay conditioning, a CS is immediately followed by an US, while in trace conditioning, CSs and USs are separated in time by a stimulus-free trace interval. Earlier studies suggested that the hippocampus (HPC) and the medial prefrontal cortex (mPFC) are recruited in trace, but not delay, fear conditioning. According to neuronal tracing studies, there is direct projection from the HPC to mPFC, but not the other way around. The Nucleus Reuniens (RE) of the midline thalamus is reciprocally connected to HPC and the mPFC and may serve as a relay station for projection from mPFC back to HPC. However, little is known regarding the role of RE in fear learning. To study this neurocircuitry, we used behavioral pharmacology approach to investigate the importance of RE in this process. We hypothesized that RE inactivation would lead to a learning deficit only in trace, but not delay, fear conditioning. Supporting our hypothesis, when we inactivated RE before conditioning, trace animals demonstrated a down-shift of fear level during retrieval test, indicating the pre-conditioning inactivation of RE impaired the acquisition of trace fear. However, inactivation of RE immediately after acquisition or before retrieval did not impair the performance of the animals. These results suggested that RE is recruited and played a critical role in acquisition, but not consolidation and retrieval, of trace fear conditioning at the behavioral level. Together, our data revealed the specific role of RE in the learning process of trace fear conditioning.