Neuroscience2014【国際学会】

2014/11/15-19にWashington DCにて開催されましたSociety for Neuroscience 2014に参加致しました．本研究室からは山本先生，大村(M2)，大西(M2)，真島(M2)，岡村(M1)，滝(M1)，大谷(M1)の7名がポスター形式で発表致しました．発表題目は以下の通りです．

Anatomical labeling with probabilistic inference for diffusion tensor imaging of human brain

Utako Yamamoto, Masayuki Ohzeki, and Tomoyuki Hiroyasu

Sex differences of the pleasantness level of pictures upon cerebral activity using fMRI

Ayumi Omura, Misato Tanaka, Utako Yamamoto, Tomoyuki Hiroyasu

Analysis of color temperature to improve intelligent productivity -The effect for psychological states and brain activity-

Natsuko Onishi, Utako Yamamoto, Mitsunori Miki, Tomoyuki Hiroyasu

Study on the related regions of working memory using reading span test

Nozomi Mashima, Utako Yamamoto, Tomoyuki Hiroyasu

Relationship between brain activity and emotional state during multi task

Tatsuya Okamura, Utako Yamamoto, Tomoyuki Hiroyasu

Analysis of processing mechanism of the human brain with multiple sensory information using fNIRS

Keinichi Taki, Utako Yamamoto, Tomoyuki Hiroyasu

3D drawing system to combine MRI data of nerve fibers and fNIRS data of activated brain regions

Shunsuke Ohtani, Utako Yamamoto, Tomoyuki Hiroyasu

学会参加報告書

報告者氏名	真島　希実
発表論文タイトル	リーディングスパンテストの高成績群と低成績群によるワーキングメモリの検討
発表論文英タイトル	Study on the related regions of working memory using reading span test of high and low score group
著者	真島希実, 山本詩子，廣安知之
主催	一般財団法人
講演会名	Neuroscienece 2014
会場	ワシントンDC
開催日程	2014/11/15-19

 
 

1. 講演会の詳細

2014/11/15-19にワシントンDCにて開催されましたSociety for Neuroscience 2014に参加致しました．この学会は神経科学の研究者や臨床医などが参加する世界最大級の脳や神経系に関連する科学的な発見を目的としています．実際に15,000以上の科学的な技術や脳の構造，健康，病気，および治療法に関する新しい研究のプレゼンテーションがなされ，数多くの革新的な研究が披露されていました．本研究室からは他に山本先生，大村さん，大西さん，岡村君，滝君，大谷君が参加しました．
 

2. 研究発表
   • 発表概要

私は学会開始2日目の8-12時のポスターセッションで参加いたしました． ４時間ポスターの前に立ち，きて下さった方に自身の研究の3分の講演時間と2分の質疑応答時間となっておりました．今回の発表は「リーディングスパンテストを用いたワーキングメモリ部位の関連の検討」で以下に抄録を記載致します．

The study of related sites of working memory using reading span test[Background and Purpose]Individual differences exist in working memory capacity. But good methods for evaluating brain functions related to working memory are unavailable. Therefore, the relationship between working memory and cerebral functions was observed using functional near-infrared spectroscopy (fNIRS). In this study, the relationship between working memory capacity and cerebral blood flow change was evaluated using the results of a reading span test (RST), which is a language working memory task.[Methods]Five healthy subjects participated in this study. Cerebral blood flow change was measured under RST using fNIRS. The regions of interest in the brain were the dorsolateral prefrontal cortex (DLPFC), Broca’s area, and Wernicke’s area which were mentioned in the Baddeley & Hitch model. It has been reported that DLPFC is responsible for continuing cautions, while Broca’s area and Wernicke’s area manages language processing. First, the functional relation between DLPFC and RST was observed. Second we evaluated association of the region in working memory using the cross-correlation function.   [Results] RST was evaluated according to the memorized number of words. Working memory capacity is larger toward the higher score. The results of the relationship between RST score and brain activities, cerebral blood flow change in DLPFC of the subjects in the high score group were greater than that of the subjects in the low score group. The task score for each subject classified them into the high score group and the low score group. According to the results of the cross-correlation function, all subjects in high score group have high coefficient of correlation value which it is over 0.7 between Broca’s area and Wernicke’s area. In contrast, low score group have high coefficients of correlation value between DLPFC and Broca’s area. [Discussions] From the result that high score subjects have large activities in DLPFC, we suggest that the brain function in DLPFC assume an important to perform RST. And high score group indicated that the language processing area has a strong association. On the other hand, low score group has a strong association between DLPFC and Broca’s area. The difference of the association area may be caused by compensation of differences in language processing. Therefore, we suggest that there may be a difference of the network system between high score group and low score group. [Conclusions] Differences were observed in cerebral blood flow change caused by the difference of the working memory capacity. In addition we suggest that the difference of the working memory capacity is caused by not only the degree of intensity but also the relation of each region.

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
 
・質問内容1
複数の外国人の方からたくさんの質問をいただきました．DTW距離のことを知らない方が多かったので，距離が小さい方が活性していうことが理解しにくかったようで，低成績の方が大きいではないかというところをたくさん聞かれました．そこはDTWの定義とどういう解析方法なのかを説明することで納得して頂けたように思います．
 
・質問内容2
明治大学の佐藤アキヒロさんのアドバイスでDTW距離を群間だけでなく，成績と距離を相関でみてみたらおもしろいのではないかというアドバイスを頂きました．
 

• 感想

私は学会開始2日目の発表ということでまだ学会の雰囲気や英語に慣れないまま本番を迎えることとなり，とても緊張しました．しかし，たくさんの方にお集まり頂き，言語の壁に苦しみつつも言いたいことを伝えられたのではないかと思います．数名の方にはinteresting! と言って頂いたことで，自分の研究に対して自信をもつことができ，さらに研究に磨きをかけていきたいと感じました．今回は最後の学会発表となりましたが，後悔なくおわることができて良かったと思います．
 

3. 聴講

今回の講演会では，下記の5件の発表を聴講しました．
 

発表タイトル ：　The activity of the mirror neuron system during being imitated by others著者 ： A. SATO, S. SHIMADAセッション名 ：  Social cognition: Behavior and pharmacologyAbstruct ： The mirror neuron system (MNS) is the brain areas that are activated when we perform some action and observe the same action performed by others. It is known that human beings tend to unconsciously imitate the behavior of others, and that the individual who is imitated tend to have good impression to the imitator, which is called the chameleon effect. Using near-infrared spectroscopy (NIRS) we investigated the activity of MNS during being imitated by the other in order to examine the relationship between MNS and the chameleon effect. Sixteen subjects participated in this experiment. The subject was instructed to write a Japanese ancient character appeared on the screen within 9s (execution phase). Then the subject watched a movie stimulus (9s) in which a model with a blue or yellow glove wrote a Japanese ancient character that is either the same or different as the subject wrote in the previous execution phase (observation phase).We measured the brain activity around C3 and C4 of the 10/20 system with NIRS throughout the experiment. After the experiment the subject evaluated how much the subject liked the blue or yellow hand in the video stimuli on a 7-point scale (1 = extremely dislikeable, 7 = extremely likeable). There was a significant difference in the hand liking score between the imitation and the control conditions : the score in the imitation condition (4.9±0.2) was significantly greater than that in the control condition (3.9±0.2, t(15)=2.18, p<.05). For the NIRS data, we performed effect size analyses for each channel in each subject and then a group-level t-test for each channel. The result showed that the activation in the left premotor cortex (ch4) was significant in both the execution and observation phases, indicating that MNS was activated when the subject was writing a character and watching the other writing a character. We then examined the correlation between the activity in the left premotor cortex (ch4) and the hand liking score. We found a significant positive correlation between the premotor activity during observation phase and the score in the imitation condition (r=0.63, p<.01). These results suggest that the more the subject likes the other who imitated the subject, the more the activity in MNS increases when the subject observed others imitating them.

この発表はミラーニューロンシスタムをfNIRSで解析した研究でした．
ミラーニューロンという自分と同じ動きをした人に好意をもつという現象において脳活動はどうなっているのかという実験でした．解析方法は加算平均処理をおこなってからeffect sizeを用いており，ミラーニューロンつまり真似された方が活性するということだった．どこで応用先を聞き逃してしまったが，とても面白い実験でした．
 
 

発表タイトル       ：Functional connectivity between player-observer motor areas during a competitive game著者                  ： S. YOKOYAMA, M. MATSUMOTO, S. SHIMADAセッション名       ： Social cognition: Behavior and pharmacologyAbstruct            ：Mirror neuron system (MNS) is the brain regions that are activated both when executing an action and when observing the other performing the same action. In this study, we measured the MNS activities of two subjects simultaneously by using 48-ch (24-ch for each subject) near-infrared spectroscopy (NIRS). Eleven pairs of right-handed male subjects participated in the experiment (aged 21.7±0.8 years). One subject (the player) performed the rock-paper-scissors (RPS) game against an experimenter and another (the observer) cheered the player playing RPS. The experiment was repeated until more than 10 trials were obtained for each outcome (win, lose, and draw). Brain activity during motor execution similar to the RPS game was also measured with both subjects (the player and the observer). We applied the general linear model (GLM) analyses on NIRS data. One-way ANOVA and its subsequent analyses on the observer’s data revealed that the inferior parietal lobule (angular gyrus) showed a significantly greater activation in the win condition compared to the draw and lose conditions (F(2,11)=4.65,P<.05). The inferior parietal lobule was also activated in the motor condition, indicating that this area is a component of the mirror neuron system. We did not find such modulation in the player’s brain activity data. We then employed the psychopysiological interaction (PPI) analysis and found that functional connectivity was observed between the premotor cortex of the player and the inferior parietal lobule of the observer in the win condition contrasted against the lose condition (t(10)=2.29, P<.05). These results indicate that the observer’s MNS was more activated and the connectivity between the player’s and the observer’s brains was more enhanced when the player whom the observer cheered won against the opponent than when he lost. Our results may account for why watching other’s performance, for example, professional sports game, is such a popular entertainment pervasively in the world.

この発表はミラーニューロンシステム時の活性部位を特定し，そのネットワークを見る研究をしていました．解析方法はまずGLMで活性部位を特定した後に，PPIという方法でネットワークをみるというものでした．GLMとネットワークを勉強していたのでこの解析方法はとても興味深く，勉強してみようと思いました．
 
 

発表タイトル       ：Differential MEG activity patterns for congruent and incongruent spatial relational judgments著者                  ： N. M. SCOTT, M. SERA, A. P. GEORGOPOULOSセッション名       ： Decision making and reasoningAbstruct            ：Children learn above/below relational judgments earlier than right/left judgments, but there is no explanation for this phenomenon. Relational judgments may exhibit congruency effects and this may complicate learning. For example, encoding above is easier at the top part of the screen than at the bottom – where it is incongruent. Incongruent trials, like right and left judgments, may require inhibition of a prepotent response or additional processing. We investigated the neural correlates of encoding and maintaining spatial relations for 3 seconds in working memory in 20 adults using magnetoencephalography (MEG). When comparing the MEG signal for above/below vs. right/left judgments over the period of stimulus presentation, we found the greatest difference in activity in the left temporal-parietal-occipital junction, right temporal cortex and cerebellum. Similar differential activation was observed during the working memory phase. When comparing congruent to incongruent trials, we found differential neural activity only in left frontal areas. Congruent-incongruent comparisons in working memory indicated differential activity bilaterally in frontal and temporal areas, including the same left frontal area indicated in the encoding phase. These findings are consistent with the idea that right/left relations are represented differently than above/below during both encoding and working memory phases.

この発表はWhite noiseの男女差の検討をMEG用いて検討していました．タスクは空間ワーキングメモリタスクで一致か不一致のときの脳活動を測定していました．この人も女性の方で、女性の方が戦略とか使えているのではと考察していました．MEGでの脳血流変化では女性は左下前頭回、男性は言語野が活性していたらしい。でもなぜ一致と不一致で比較する必要あったのかについては理解できませんでした。
 

発表タイトル       ：Imagine the future! How does episodic simulation enhance prospective memory?著者                  ： I. MOMENNEJAD, J. D. COHEN, K. A. NORMANセッション名       ： Human long-term memory: Encoding-retrieval interactionsAbstruct            ：　Imagining task-related future episodes enhances prospective memory (PM) (Brewer et al. 2011). One hypothesis is that episodic simulation recruits mechanisms that spontaneously bring prospective intentions to mind at the right time (Einstein and McDaniel, 2000). However, the neural mechanisms underlying the role of episodic future simulation in prospective memory have not yet been established. We suggest that (a) spontaneous retrieval processes in PM rely on episodic memory (Cohen and O’Reilly, 1996), and (b) imagining task-related future episodes (episodic “preplay”) increases the odds that the memory trace laid down during intention-formation will match the participant’s mental state when the PM target appears. This increase in study-test match should (in turn) increase the likelihood that the PM target will trigger spontaneous retrieval of the PM intention. This view predicts that neural evidence for accurate episodic future simulation (i.e., preplay of features that match the PM target) would correlate with prospective memory success. To test this prediction, we used multi-voxel pattern analysis (MVPA) of functional magnetic resonance imaging (fMRI) data from a non-focal prospective memory paradigm. Participants were instructed to perform a numerical parity or magnitude task at fixation (ongoing task). while faces and scenes appeared peripherally on the left or right side of the screen. Participants were instructed to respond to a specific peripheral face or scene image (non-focal PM target) when it appeared on a specific side of the screen (left or right). In each trial, participants were informed about their future target (the identity of the image, and the side of the screen on which it would appear) and the ongoing task they would be performing when it appeared. After instruction, participants were asked to mentally simulate the future target and rate the quality of their imagery (20 s) before the start of the task. Behaviorally, we found a positive correlation between the ratings of imagery quality and PM hit rate. In order to test whether neural evidence for future simulation correlated with PM success, we trained a classifier on a functional localizer with images of faces and scenes (appearing on the left or right, during an ongoing parity or magnitude task), and we used these parameters to decode what the subjects “imagined” during the simulation period, along the following four dimensions: left/right, image category, specific image identity, and ongoing task (parity/magnitude). We will present the results of analyses exploring whether MVPA evidence of preplay during the simulation period predicts PM success.

この発表はエピソードバッファとprospective memory の関係を研究発表内容でした。
エピソードバッファはワーキングメモリの中でもわかっていないことが多いので、興味ももち聞きに行ってきましたが、prospective memory とはワーキングメモリとは違うものであることがわかりました。あるゴールを目標としたときのそれに向かう継続的な行動のことを指すようです。課題もとても複雑そうで理解できなかったのでまた調べてみたいと思います。
 
 

発表タイトル       ：Direct recordings of sustained theta-band electrical activity in the human auditory cortex during working memory for tones著者                  ： P. E. GANDER, S. KUMAR, K. V. NOURSKI,H. OYA, H. KAWASAKI, M. A. HOWARD, T. D. GRIFFITHSセッション名       ： Working memoryAbstruct            ：　Working memory is the capacity to hold and manipulate behaviourally relevant information in mind in the absence of ongoing sensory input. Here we tested the hypothesis that auditory working memory for tones requires ongoing activity in auditory cortex, and examined the form of such activity in neuronal ensembles. We recorded local field potentials from two human subjects undergoing invasive monitoring for pre-surgical localization of epileptic foci. The subjects were implanted with depth electrodes along the axis of Heschl’s gyrus (HG) containing primary cortex in the medial part, and subdural grids over temporal and frontal cortex. Following a visual alert subjects were presented with a pair of tones (0.5 s duration, 1 s ISI) belonging to two different categories (‘Low’: 300-570 Hz; ‘High’: 2000 -2800 Hz). A visual cue (1.5 s) then informed the subjects which tone (first or second) to keep in mind. A 3 s retention period was followed by a tone which could be the same or different (frequency difference ±20%) from the tone held in mind. The subjects made a same/different decision by pressing a button. A total of 96 trials (48 each of ‘Low’ and ‘High’ tone retention) were presented. We measured average ERPs and carried out single-trial time-frequency analysis using a wavelet transform. Both the magnitude of ERPs (~100 ms after stimulus onset) and gamma-band (60-120 Hz) power in electrodes located in HG and lateral superior temporal gyrus (STG) showed category-specific responses during the perception of tones. High tones elicited stronger responses in medial HG and low tones in lateral HG. During the retention period, a sustained theta-band (2-6 Hz) activity was observed in all contacts that showed gamma-band responses during perception. Power in this band also showed a recency effect: a greater response in HG electrodes was observed for retention of the most recently presented (second) tone. On the STG, however, the opposite effect was observed: a greater 2-6 Hz power for retention of the first compared to the second tone. The data demonstrate: 1) theta-band correlates of tone retention in auditory cortex in the same neural ensembles that are active in the gamma band during perception 2) neural bases in the auditory cortex for interference effects within tonal working memory.

この発表は音程と聴覚ワーキングメモリの関係をMRIとERPを用いて測定した研究でした。
空間ワーキングメモリと言語ワーキングメモリは有名ですが聴覚ワーキングメモリをはじめて知りました。結果として、色調のワーキングメモリの干渉効果は聴覚皮質の神経と相関あったそうです。
ワーキングメモリといっても概念が広く，感覚野の関わりも違ってくることを改めて感じました。
 
参考文献

• Society for Neuroscience 2014, http://www.sfn.org/

 
学会参加報告書

報告者氏名	大谷俊介
発表論文タイトル	DTIとfNIRSデータを用いた脳活性部位間の神経線維3D描画システムの構築
発表論文英タイトル	3D drawing system to combine DTI data of nerve fibers and fNIRS data of activated brain regions
著者	大谷俊介，山本詩子，廣安知之
主催	Neuroscience
講演会名	Neuroscience2014
会場	Convension Center
開催日程	2014/11/15-2014/11/19

 
 

1. 講演会の詳細

2014/11/15から2014/11/19にかけて，Washington,DCのConvension Centerにて開催されましたNeuroscience2014に参加いたしました．このNeuroscience2014は，Neuroscienceよって主催された研究会で，世界的な神経科学のためのアイデアやツールを紹介し，意見を交換し合うことを目的に開催されました．
私は15，16，17，18，19日の5日間参加いたしました．本研究室からは他に山本先生，真島さん，大西さん，大村さん，岡村，滝が参加しました．
 

2. 研究発表
   • 発表概要

私は16日の午前のセッション「Imaging Advances: Silico and High Throughput」に参加いたしました．発表の形式はポスター発表で，4時間の発表時間となっておりました．
今回の発表は，3D drawing system to combine DTI data of nerve fibers and fNIRS data of activated brain regionという題で発表しました．発表内容は提案システム用いることで，DTIとfNIRSデータで，機能的に繋がりが示唆された領域間に構造的な繋がりの存在が確認できるというものです．以下に抄録を記載致します．

抄録中身Objective We aim to elucidate the brain network that leads to clarification of the pathogenic mechanism of mental illness. As fundamental examination, we visualized in three dimensions the brain active regions obtained functional Near-Infrared Spectroscopy (fNIRS) and the running of the nerve fibers of brain obtained by nerve tracking method using Diffusion Tensor Imaging (DTI) data of Magnetic Resonance Imaging (MRI) with Direct3D, which is a kind of 3D programming. We discuss to study the relationship of the nerve fibers in the brain between the active regions using the proposed system. Methods In order to visualize the connection of the nerve fibers, the proposed system drew the nerve fibers between the active regions with three-dimensional coordinate data of the nerve fibers obtained from neural tracking method using DTI data，active CH determined from cerebral blood flow change data of fNIRS and three-dimensional coordinate data of each of CH obtained from the magnetic measurement of fNIRS. Experiments using this system discussed the relationship of several active regions when subjects perform RST task that store a designated word in a sentence during reading the short sentence. A previous study has reported that a person with low scores had activation in the Broker region and Dorsolateral prefrontal cortex(DLPFC). On the other hand, a person with high scores had activation in the Broker region．DLPFC and the Wernicke’s region. In this study, one person with high scores and one person low scores participated. We drew the nerve fibers between the active regions of each subject by our system. Compared with results of each subject, and we investigated the relationship of the nerve fibers between the active regions. Results In result of using this system, it was confirmed that the low-score subject had intense connection of nerve fibers between the Broker region and DLPFC. On the other hand, high-score subject had connections of nerve fibers Broker region and DLPFC, the Broker region and the Wernicke’s region. These results indicated that the structure of the connection of the nerve fibers in the brain between the specific regions had the difference, and it depends on the results of RST. Conclusions In comparison with subjects whose results are different, the difference was found inThe connection of the nerve fibers. Connections of nerve fibers presented the connections in the regions suggested by several previous studies. We could evaluate the nerve fibers between the active regions using fNIRS data and MRI data by the proposed system. Visualizing the nerve fibers of the active regions using this system suggested that could lead to the elucidation of brain network.

 
 

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
 
 
・質問内容1
質問内容は，「神経線維はどのように描画したのか」というものでした．この質問に対する私の回答は，「DTIを撮像し，3Dプログラミングの一つのDirect3Dを用いて描画した」というものでした．
・質問内容2
質問内容は，「MRIとNIRSの原点をどのように合わせたのか」というものでした．この質問に対する私の回答は，「それぞれの原点を求めて，目視で合わせた」というものでした．
・質問内容3
質問内容は，「DTIデータは一点のデータに対してfNIRSデータは時系列データであるのに，
どのようにして対応させたのか」というものでした．この質問に対する私の回答は，「時系列データのある時間に対応する，DTI構造データを抽出する」というものでした．
・質問内容4
質問内容は，「あなたのキャリアはコンピュータサイエンスなのか生命科学なのか」というものでした．この質問に対する私の回答は，「コンピュータサイエンス」というものでした．
 

• 感想

発表時間が長いこともあって，外国人，日本人ともに多くの人に自分の研究内容をしっかり伝えることが出来た．しかし，質問の意味がわかり，うまく答えられるものもあれば，どう答えたらいいのかわからないものもあった．英語力のなさを痛感した学会でした．でも，研究内容についてはおもしろい，興味があると言ってもらえたので良かったと思います．もっと議論ができると，なおさら面白い学会になったと思うので，英語を頑張ろうと思いました．
 

3. 聴講

今回の講演会では，下記の5件の発表を聴講しました．
 

発表タイトル       ：　The activity of the mirror neuron system during being imitated by others著者                  ： A. SATO， S. SHIMADAセッション名       ： Social cognition: Behavior and pharmacologyAbstruct            ：The mirror neuron system (MNS) is the brain areas that are activated when we perform some action and observe the same action performed by others. It is known that human beings tend to unconsciously imitate the behavior of others, and that the individual who is imitated tend to have good impression to the imitator, which is called the chameleon effect. Using near-infrared spectroscopy (NIRS) we investigated the activity of MNS during being imitated by the other in order to examine the relationship between MNS and the chameleon effect. Sixteen subjects participated in this experiment. The subject was instructed to write a Japanese ancient character appeared on the screen within 9s (execution phase). Then the subject watched a movie stimulus (9s) in which a model with a blue or yellow glove wrote a Japanese ancient character that is either the same or different as the subject wrote in the previous execution phase (observation phase).We measured the brain activity around C3 and C4 of the 10/20 system with NIRS throughout the experiment. After the experiment the subject evaluated how much the subject liked the blue or yellow hand in the video stimuli on a 7-point scale (1 = extremely dislikeable, 7 = extremely likeable). There was a significant difference in the hand liking score between the imitation and the control conditions :the score in the imitation condition (4.9±0.2) was significantly greater than that in the control condition (3.9±0.2, t(15)=2.18, p<.05). For the NIRS data, we performed effect size analyses for each channel in each subject and then a group-level t-test for each channel. The result showed that the activation in the left premotor cortex (ch4) was significant in both the execution and observation phases, indicating that MNS was activated when the subject was writing a character and watching the other writing a character. We then examined the correlation between the activity in the left premotor cortex (ch4) and the hand liking score. We found a significant positive correlation between the premotor activity during observation phase and the score in the imitation condition (r=0.63, p<.01). These results suggest that the more the subject likes the other who imitated the subject, the more the activity in MNS increases when the subject observed others imitating them.

 
この研究は，fNIRS装置を用いてミラーニューロンシステムとカメレオン効果の関係について検討している．このカメレオン効果とは，相手の行動を意識していないのにも関わらず，行動をまねるという効果である．これは，協調となにか関係があるのではないかと考えました．協調作業では下前頭回がもっとも関連する部位と言われているが，この研究のように運動野を着目したらどうなるのかなと感じました．
 

発表タイトル       ：Transcriptome tomography: Mapping genes onto 3D brain structures著者                  ：Y. OKAMURA-OHOセッション名       ： Genomics, proteomics, and systems biologyAbstruct            ：Expression-anatomy association is crucial for understanding molecular functions of, in particular, novel coding and non-coding genes. We have invented a framework for comprehensive gene expression density mapping on the whole three dimensional (3D) anatomical context, Transcriptome Tomography (PLoS One,2012;7,e45373,Video http://www.youtube.com/watch?v=Td4rGRQIZuY&list=UUIGmhpdcVev1Wc0YK7FHIig). Expression densities measured with high-throughput methods are usable for their mapping onto the 3D brain structure detected with other modalities such as MRI. Also they are usable directly for gene-by-gene correlation analysis of co-expression. This framework comprehensively assesses co-expression patterns that are latent within expression maps. Expression maps and co-expression search results in the mouse brain can be browsed in our website, ViBrism-DB (http://vibrism.neuroinf.jp/). In this presentation we would focus on previously uncharacterized mouse-specific genes that were co-expressed with gene groups encoding transcription factors and related molecules. The genes were expressed in areas associated with specific brain functions and the previously uncharacterized non-coding genes were located in a co-expression network position linking the groups. This linkage suggests characteristics of the non-coding genes that may coordinate multiple gene groups and create mouse-specific neural designs. A part of this work was supported by members in RIKEN Neuroinformatics Japan Center and conducted within the WHS and DAI Task Forces of the INCF Program on Digital Brain Atlasing.

 
本発表は，遺伝子情報を脳構造にマッピングするという研究である．遺伝情報の位置情報はどのようにしているのか気になったが，この研究が進むと，人の疾患についてモデル動物での研究を進めることが可能になる．そして，網羅的な発現情報を3次元空間にマッピングすることで，創薬などのターゲット分子の探索ができるのではないかと考えた．
 

発表タイトル       ：Functional connectivity between player-observer motor areas during a competitive game著者                  ：S. YOKOYAMAセッション名       ： Social cognition: Behavior and pharmacologyAbstruct            ：Mirror neuron system (MNS) is the brain regions that are activated both when executing an action and when observing the other performing the same action. In this study, we measured the MNS activities of two subjects simultaneously by using 48-ch (24-ch for each subject) near-infrared spectroscopy (NIRS). Eleven pairs of right-handed male subjects participated in the experiment (aged 21.7±0.8 years). One subject (the player) performed the rock-paper-scissors (RPS) game against an experimenter and another (the observer) cheered the player playing RPS. The experiment was repeated until more than 10 trials were obtained for each outcome (win, lose, and draw). Brain activity during motor execution similar to the RPS game was also measured with both subjects (the player and the observer). We applied the general linear model (GLM) analyses on NIRS data. One-way ANOVA and its subsequent analyses on the observer’s data revealed that the inferior parietal lobule (angular gyrus) showed a significantly greater activation in the win condition compared to the draw and lose conditions (F(2,11)=4.65,P<.05). The inferior parietal lobule was also activated in the motor condition, indicating that this area is a component of the mirror neuron system. We did not find such modulation in the player’s brain activity data. We then employed the psychopysiological interaction (PPI) analysis and found that functional connectivity was observed between the premotor cortex of the player and the inferior parietal lobule of the observer in the win condition contrasted against the lose condition (t(10)=2.29, P<.05). These results indicate that the observer’s MNS was more activated and the connectivity between the player’s and the observer’s brains was more enhanced when the player whom the observer cheered won against the opponent than when he lost. Our results may account for why watching other’s performance, for example, professional sports game, is such a popular entertainment pervasively in the world.

 
本発表は，じゃんけんをする人とそれを見ている人の脳機能の関係を調査している研究である．勝利した人をみている人のミラーニューロンシステム(MNS)がより活性したという結果が得られていた．スポーツ観戦で勝てばうれしい気持ちになるのは，このMNSの活性が関係しているのではないかと考えた．MNSの活性に伴い，なんらかの神経伝達物質が出ていると思うので，本研究で物質もみていけば，よりMNSの解明に繋がると感じた．
 
 

発表タイトル       ：Reward- and conflict-induced interference: Guiding and biasing of visual attention著者                  ：C. GIATTINOセッション名       ： Functional mechanisms of attentionAbstruct            ： Features in the world around us are constantly biasing our attention. When these features are associated with reward (and are thus more salient), they can bias our attention, thereby promoting – or sometimes impairing – cognitive task performance, depending on the relevance of those features for our goals. Here we examined how visual attention and conflict processing interact with reward associations. Participants performed a version of the Erikson flanker paradigm in which the central target letter could be surrounded by symmetric flankers (e.g., XXYXX [incongruent] or YYYYY [congruent]) or asymmetric ones (e.g., XXYYY, left congruent, right incongruent), while neurophysiological (EEG) data was recorded, from which we extracted time-locked event-related potential (ERP) and event-related spectral (ERSP) responses. Additionally, two of a set of four letters were associated with high reward and the other two with low reward, all of which could act either as targets or flankers in different trials. Participants could gain money depending on their response time (RT) performance. Results showed that RTs increased as the number of incongruent flankers increased (bilateral congruent RTs < asymmetric incongruent RTs < bilateral incongruent RTs), as well as when the target letter was associated with high reward compared to low reward. In addition, if flanking incongruent letters were associated with high (vs. low) reward, participants responded more slowly, suggesting a reward-induced increase in attentional distraction. Asymmetrical conflict effects were observed in the ERPs with a lateralized negative deflection (at ~250-400 ms), followed by an increase in oscillatory power in the alpha band (8-12 Hz) substantially later (at ~600-900 ms), suggesting an initial distraction-induced attentional orienting toward the conflicting letters, followed by an active suppression of that input. In addition, preliminary analyses indicated that bilateral high-reward flankers elicited enhanced attentional biasing, as reflected by a decrease in occipital alpha. Finally, the analyses suggested that high-reward targets (irrespective of flanker reward) increased attentional biasing activity frontocentrally, manifested by a decrease in oscillatory beta-band activity (12-18 Hz). Together, these results reveal the different mechanisms by which reward can guide attention to suppress the processing of irrelevant distracting features (e.g., incongruent flankers) and to enhance the processing of relevant features (e.g., reward-associated targets) via top-down biasing

 
この研究では，課題の反応時間によってお金をあげるという報酬を与えることによって，注意機能はどうなるかを調査する研究である．報酬の額によっても反応時間が異なるという結果も得られている．注意機能は意欲によって左右されるという結果なので，注意機能を調査する際には，被験者の意識をそろえる必要があると感じた．
 
 
 

発表タイトル       ：DTI analysis of corticospinal tract using BrainSuite: A potential biomarker of upper extremity therapeutic response to neurorehabilitation in chronic stroke著者                  ：B. KIMセッション名       ： Stroke imaging and diagnostic studiesAbstruct            ：Corticospinal tract (CST) microstructural characteristics measured by diffusion tensor imaging (DTI) are known to be associated with upper extremity (UE) motor impairment after stroke. However, there is a gap in understanding the relationship between DTI-derived measures and UE motor function changes following neurorehabilitation. This study is part of a larger longitudinal phase-I clinical trial in chronic stroke that aims to determine the optimal dose of therapy for sustained affected arm use after therapy has ended. Our purpose is twofold: First, to establish methods to quantify CST characteristics in lesioned brains using BrainSuite. Second, to determine if UE motor performance changes after treatment in chronic stroke are associated with DTI-based CST measures. Nine chronic stroke participants completed neuroimaging and clinical assessments before and after 12 sessions of a reproducible UE therapy program within 4 months. Imaging data were processed using BrainSuite (version 13a, http://brainsuite.org). Specifically, BrainSuite was used to semi-automatically extract and parcellate the participants’ brains from T1-weighted structural MRI images, to correct the diffusion images for geometric distortion, to coregister the diffusion images with the structural images, to compute DTI parameters, to perform deterministic tractography, and to identify the CST based on the set of tracks that pass through both a manually labeled pons region of interest (ROI) and an automatically labeled precentral gyrus ROI. The ipsi- and contra-lesional CST fractional anisotropy (FA) was quantified, and the CST FA asymmetry index [(FAcontra – FAipsi)/(FAcontra + FAipsi)] was calculated. The Wolf motor function test (WMFT) and Fugl-Meyer assessment (FMA) were performed to assess participants’ motor function and impairment respectively. Linear regression analysis was performed to examine the relationship between CST FA asymmetry and these clinical outcomes, based on the hypothesis that pre CST FA asymmetry would be correlated with the pre to post changes in the WMFT and FMA scores. Similar to previous reports, the pre CST FA asymmetry was positively correlated with the pre WMFT score. However, the pre and pre to post changes in CST FA asymmetry were not correlated with the pre FMA or pre to post changes in either FMA or WMFT scores. Because of the limited range of pre CST FA asymmetry indices for the nine participants in this study, a relationship between CST FA asymmetry and clinical outcomes could not be explained by our results. Thus, further investigations of DTI-derived measures in post-stroke individuals are necessary to identify biomarkers for functional recovery.

 
本発表では，脳卒中後の運動機能の変化をDTIを用いて評価することを目的にしている研究である．精神疾患の評価などではよくDTIが用いられているが，神経リハビリテーション関係にDTIを用いられていることはあまり聞かない．FA値の比較を行うことで神経の髄鞘化を検討することができると思うでは，リハビリの分野でもDTIを用いた評価ができることが可能だと感じた．
 
学会参加報告書

報告者氏名	大村歩
発表論文タイトル	Sex differences of the pleasantness level of pictures upon cerebral activity using fMRI
発表論文英タイトル	Sex differences of the pleasantness level of pictures upon cerebral activity using fMRI
著者	A. OMURA, M. TANAKA, U. YAMAMOTO, T. HIROYASU
主催	Society for Neuroscience
講演会名	Neuroscience2014
会場	Washington Convention Center
開催日程	2014/11/15-2014/11/19

 
 

1. 講演会の詳細

2014/11/15から2014/11/19にかけて，Washington Convention Centerにて開催されましたNeuroscience2014に参加いたしました．このNeuroscience2014は，Society for Neuroscienceによって主催された学会で，臨床・基礎・技術系の多様な参加者による自由で闊達な交流と議論の場となることを目的に開催されています．
私は学会期間の全日参加いたしました．本研究室からは他に山本先生，真島さん，大西さん，岡村くん，大谷くん，滝くんが参加されました．
 

2. 研究発表
   • 発表概要

私は19日の午前のポスターセッション「Human Emotion: Individual Differences and Disorders」に参加いたしました．発表の形式はポスター発表で，セッション時間(8:00-12:00)，特に指定された時間(9:00-10:00)の間に来て下さった方に発表するという形式でした．
今回の発表は，快の度合が異なる際の男女における脳活動の違いの検討について発表しました．以下に抄録を記載致します．

抄録中身[Purpose]The goal of this study was to identify an indicator of the level of pleasantness based on information provided by functional brain mapping. Because it is considered that work of emotion changes with factors of age and sex, the purpose was to verify differences in regions of brain activity and the level of activation owing to sex.[Methods]In this study, using a visual experiment whose stimulus pictures have the level of pleasantness, we considered sex differences in regions of brain activity and the level of activation based on the levels of pleasantness. Fifteen healthy right-handed subjects (8 women and 7 men) participated in this study. The group had a mean age of 21.2 years (standard deviation, 0.4 years). In this study, the level of pleasantness was categorized into three levels: High, Medium, and Neutral. Stimulus pictures which were sorted by the levels of pleasantness were presented to the subjects, and we measured Blood Oxygenation Level Dependent (BOLD) signal using Magnetic Resonance Imaging (MRI). Obtained data was analyzed by group analysis. Subsequently, we examined the differences in regions of brain activity and the level of activation owing to sex. [Results] Compared with men, women tended to activate in the cingulate cortex and the frontal lobe. Only in the case of High, women had activation in the cingulate cortex. In the case of Medium and Neutral, they did not have activation in the cingulate cortex. Besides, as the level of pleasantness decreased, the activation in the frontal lobe became smaller. On the other hand, men did not have activation in the cingulate cortex in the case of High. As with women, a lower level of pleasantness was associated with a smaller activation in the frontal lobe However, compared with women, they had a tendency for activation in the frontal lobe to be small. The frontal lobe is related to attention and interest. As a higher level of pleasantness induces higher attention and interest, it is considered that a higher level of pleasantness yields a larger active region. The level of pleasantness in sex is considered to reflect in the frontal lobe as the extent of activity, because a previous study has reported that women had more intense emotion than men did. [Conclusions] In this study, we investigated the differences in regions of brain activity and the level of activation owing to sex. In both men and women, a lower level of pleasantness was associated with smaller activated areas in the frontal lobe. Furthermore, men had lower activation than women. Consequently, it is suggested that level of pleasantness and sex differences could be identified.

 

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
 
・質問内容1
脳の賦活で，左右差はないのか?右の方がより賦活しているのかな?という質問を頂きました．これに対しては，左右差はあまり検討していないが，右の方が総じて賦活している傾向にある，と回答しました．
 
・質問内容2
画像をもっと本能に働きかけるような画像(たとえば食べ物のみなど)に偏らせてはどうか?という助言を頂きました．これに対しては，様々なパターンの快に対応したいので今は様々なテーマの画像を使用しているが，検討してみます，と回答しました．
 
・質問内容3
各快の度合でこの男女差だったのか(つまり海馬での差が見られたのか)?という質問を頂きました．これに対しては，はい，海馬で各度合で男女で差分が見られました，と回答しました．
 
・質問内容4
Unpleasantは使ってないのか?という質問を頂きました．これに対しては今回はPleasantのみで，Unpleasantは使用していません，と回答しました．
 

• 感想

英語で初のポスター発表で，発表の長さも決まっておらず，非常に緊張しました．ポスターを見に来て頂いた方が何をおっしゃっているのか分からない場合も多く，非常に申し訳ない気持ちとともに，自分の英語力の無さに悔しくなりました．来て下さった方はあまり多くありませんでしたが，少しでも英語で研究についてディスカッションでき，非常に有意義なものだったと思います．
 

3. 聴講

今回の講演会では，下記の5つの発表を聴講しました．
 

発表タイトル       ：　Differential contribution of the medial and lateral entorhinal cortices, the perirhinal and postrhinal cortices and the hippocampal subfields to familiarity著者                  ： A. MALET-KARAS, N. NAKAMURA, T. KITSUKAWA, M. SAUVAGEセッション名       ： Learning and memory: Cortical and hippocampal circuitsAbstruct            ： Recognition memory is known to rely on two distinct processes recollection and familiarity. It is well-accepted that the hippocampus supports recollection. However, the specific neural basis for familiarity remains elusive. Indeed, a major controversy in recognition memory is whether only the parahippocampal region supports familiarity or whether the hippocampus also does in addition to recollection. This discrepancy emerged principally because of the lack of tools with spatial resolution high enough to dissociate activity occurring in adjacent regions in humans (for example the hippocampus and the parahippocampal region). Also, a growing number studies have reported a functional segregation within the parahippocampal region itself, e.g. between the lateral and medial entorhinal cortices and the perirhinal cortex and postrhinal cortices. Yet, the specific contribution of those areas to familiarity remains unclear. Moreover, the medial entorhinal cortex was recently found to selectively contribute to recollection and not familiarity, suggesting that not all parahippocampal areas would support familiarity. Here, we tested this hypothesis by combining a rat behavioral memory paradigm shown to yield familiarity-only judgments and high resolution molecular imaging to map the activity of the different MTL areas (the lateral and medial entorhinal cortices, the perirhinal cortex and postrhinal cortices, CA1 and CA3) during the retrieval of odor memory. The neuronal activation is assessed by the detection by fluorescent in-situ hybridization of the immediate-early gene Arc which is tightly linked to plasticity processes. Preliminary results indicate a functional segregation of the MTL areas in terms of their contribution to the familiarity process.

この発表は記憶のプロセスについて検討している研究でした．自分の研究には記憶はあまり関係ありませんが，海馬は情動記憶を担っている部位で，自分にも関係のある部位だったので非常に勉強になりました．
 

発表タイトル       ：Conscious level required for working memory and long-term memory functions: An anesthesia study in humans著者                  ： M. OSAKA, K. ENDO, T. IKEDA, T. MINAMOTO, A. NAKAE, S. HAGIHIRA, Y. FUJINO, T. MASHIMOセッション名       ： ++F.01.o. Working memoryAbstruct            ： Anesthesia is known to impair large variety of cognitive functions, disturbing neural synchronization. However, there are few studies systematically investigating an effect of anesthetic drugs on those functions. Manipulating a level of sedation, the present study examined the effect of anesthesia on two main cognitive functions: working memory and long-term memory. Participants received either Propofol (n = 10) or Midazolam (n = 10) and their cognitive performance was measured in four different sedative phases: before anesthesia (baseline), deep, moderate, and light phases. The sedation phases were manipulated individually with a target control infusion. We measured a reference level that produced conscious loss based on an eyelash reflex, and decreased the level in 1/4 step for each phase: 3/4 level for the deep, 2/4 for the moderate, and 1/4 for the light. Working memory was measured with a dual task paradigm where participants were required to remember words in a target category while ignoring non-category words. Following the task, a recognition task for the target words was given in order to test long-term memory. Another recognition task was given after the experiment when conscious level was fully recovered. Five participants were removed due to extremely poor performance (3 SDs away from the mean). In the working memory task, performance in the deepest phase was lower in comparison to the baseline and light conditions across two drugs. The result indicates that more than half of the conscious level was required for normal working memory function. On the other hand, in the first recognition task, performance in the deep condition was lower than those in other three conditions, indicating that half of the conscious level allows for normal long-term memory. Interestingly, in the second recognition task, performance in the moderate sedation was significantly lower than those in the baseline and light sedation. Those results indicate that memory transference is required three-quarter of the consciousness.

この発表は麻酔状態のときにワーキングメモリ課題を行い，意識レベルの違いでレスポンスがどう異なるかについて検討した研究でした．麻酔状態下におけるワーキングメモリという珍しくて面白い検討を行っており，より医療現場の臨床に近い実験だったので非常に興味深かったです．
 

発表タイトル       ：　The activity of the mirror neuron system during being imitated by others著者                  ： A. SATO, S. SHIMADAセッション名       ： ++F.01.u. Social cognition: Behavior and pharmacologyAbstruct ： The mirror neuron system (MNS) is the brain areas that are activated when we perform some action and observe the same action performed by others. It is known that human beings tend to unconsciously imitate the behavior of others, and that the individual who is imitated tend to have good impression to the imitator, which is called the chameleon effect. Using near-infrared spectroscopy (NIRS) we investigated the activity of MNS during being imitated by the other in order to examine the relationship between MNS and the chameleon effect. Sixteen subjects participated in this experiment. The subject was instructed to write a Japanese ancient character appeared on the screen within 9s (execution phase). Then the subject watched a movie stimulus (9s) in which a model with a blue or yellow glove wrote a Japanese ancient character that is either the same or different as the subject wrote in the previous execution phase (observation phase).We measured the brain activity around C3 and C4 of the 10/20 system with NIRS throughout the experiment. After the experiment the subject evaluated how much the subject liked the blue or yellow hand in the video stimuli on a 7-point scale (1 = extremely dislikeable, 7 = extremely likeable). There was a significant difference in the hand liking score between the imitation and the control conditions : the score in the imitation condition (4.9±0.2) was significantly greater than that in the control condition (3.9±0.2, t(15)=2.18, p<.05). For the NIRS data, we performed effect size analyses for each channel in each subject and then a group-level t-test for each channel. The result showed that the activation in the left premotor cortex (ch4) was significant in both the execution and observation phases, indicating that MNS was activated when the subject was writing a character and watching the other writing a character. We then examined the correlation between the activity in the left premotor cortex (ch4) and the hand liking score. We found a significant positive correlation between the premotor activity during observation phase and the score in the imitation condition (r=0.63, p<.01). These results suggest that the more the subject likes the other who imitated the subject, the more the activity in MNS increases when the subject observed others imitating them.

この発表は書いた文字をまねされた時のミラーニューロンの賦活についてfNIRSを用いて検討した発表でした．まねされなかったときよりもまねされた時の方が，CH4(Premotor cortex)がより賦活するという結果で，やはりまねされると人は心地よく感じるのだなと感心しました．心地よさという点では自分の研究に関係があり，大変興味深かったです．
 
 

発表タイトル       ：　Developing support vector machine classification of associative memory for real-time fMRI著者                  ： H. DESHPANDE, A. EKLUND, J. LISINSKI, C. MUELLER, B. KING-CASAS, S. LACONTEセッション名       ： ++F.01.h. Human long-term memory: Encoding-retrieval interactionsAbstruct ： Introduction: Classification-based real-time functional magnetic resonance imaging (rtfMRI) can decode subjects’ brain activity to control an fMRI stimulus, potentially enabling therapeutic neurofeedback. Based on the success of region-of-interest-based rtfMRI [1], we are developing a support vector machine (SVM)-based rtfMRI system that can explore various facets of learning and memory. Reported here are results from a verbal paired associate memory study. Methods: fMRI data were collected on a 3T MRI (TR/TE = 2000/30 ms, 64×64 acquisition matrix, 3.6-mm slice thickness, 33 slices, 220-mm field of view). Participants first performed a memorization task with 70 pseudo-word pairs [2] followed by a multiple choice recognition task. They performed 2 runs of each task in the scanner, with identical word pairs presented in random orders and with different multiple choice distractors. Preprocessing and SVM classification were performed using AFNI [3] and 3dsvm [4]. The memorization data were “labeled” with their subsequent recognition performance to train and test an SVM. By alternatively using one memorization run as training data and the other as test data, we were able to obtain cross-validated classification accuracies. Results: The subjects’ recognition performance on the first and second recognition tasks was 40% and 50%, respectively (chance = 25%). The average SVM accuracy was 53% (31% and 22% were correctly classified as learned and not learned, respectively. While 19% and 27% were incorrectly predicted as learned and not learned, respectively). Fig. 1 shows examples of actual SVM output during correct fMRI classifications. Conclusions: Our results support the possibility of designing rtfMRI experiments to develop refined learning strategies as well as to neurofeedback-based rehabilitation for memory deficits after stroke/head injuries. References: [1] Gabrieli, et al. 2012. [2] Medler and Binder, 2005. [3] Cox, 1996. [4] LaConte, et al. 2005.

この発表はリアルタイムfMRIで得られたデータをSVMを用いて識別するという内容でした．fMRIデータを識別している点，また，リアルタイム処理という点が非常に興味を引く点でした．リアルタイムということで識別に用いる特徴量を選ぶときにROIを設定してらず，全脳のボクセルを特徴量に用いているそうで，それで識別できるのか少し疑問に感じました．
 
 

発表タイトル       ：　Involvement of bilateral inferior frontal gyri in emotional recognition with social context: A TMS study著者                  ： Y. YAMASHITA, H. MAESHIMA, M. ABE, M. HONDA, M. OKADA, K. OKANOYAセッション名       ： ++F.01.n. EmotionAbstruct ：In order to investigate the involvement of both left and right inferior frontal gyri (IFG) in emotional recognition with social context, we developed a novel behavioral task to quantitatively measure an effect of contextual information on the recognition of facial expression changes. Then, by applying repetitive transcranial magnetic stimulation (rTMS) over left and right IFG, we examined whether these brain regions are critically involved in the processing of emotional context. The task trial consisted of sequential presentation of three pictures of a pre-face, a context scene, and a post-face, selected from the facial expression continua that were morphed in 6 steps from happy to fearful faces of a same model identity. Inserted scene images were pleasant, unpleasant scenes describing social context, and meaningless images were used for control conditions. The participants were asked to discriminate quickly and accurately the direction of facial expression changes (positive/negative) from the pre-face to the post-face. The behavioral experiment demonstrated that the detection of subtle facial expression change was highly sensitive to the context scene. That is, when a facial expression change was “compatible” (i.e. direction of a change in facial expression was consistent with the context stimulus), subjects were able to recognize significantly more precisely than “incompatible” changes of facial expressions (context effect). In addition, when pairs of the pre and the post faces belonged to different emotional category, subjects were able to recognize significantly more quickly and precisely than those from same category (category effect). Eighteen subjects, in whom the left IFG was successfully identified using TMS-induced speech arrest (SA) method, participated in the rTMS experiment. Fifteen minute-rTMS of 1Hz was applied to the left and right IFG to interfere cortical functions of these areas and the performances of the above task were measured before, immediately after and 30 minutes after the rTMS. As a result, the context effect was affected by the TMS applied over both the right and the left IFG. That is, TMS temporarily increased the error rate for the contextual trials, but not for the control, suggesting that bilateral IFG may be critically involved in the processing of emotional context. By contrast, the category effect was selectively affected by the TMS applied over the left IFG, which increased response latencies for the different category trials but not for the same category trials. This suggests that the left IFG plays a critical role in the categorical perception of emotional facial expression.

この発表は書いた感情認知をTMSを用いて検討していました．TMSについて詳しく知れたのはこの発表を聞いてからだったので，大変勉強になりました．最小限の不快感で脳活動を引き起こさせるTMSは，快感情を研究している自分にとっては関係のあるトピックで面白かったです．
 
学会参加報告書

報告者氏名	岡村達也
発表論文タイトル
発表論文英タイトル	Relationship between brain activity and emotional state during multi-task
著者	岡村達也，山本詩子，廣安知之
主催	Society for Neuroscience
講演会名	Neuroscience2014
会場	Walter E. Washington Convention Center
開催日程	2014/11/15-2014/11/19

 
 

1. 講演会の詳細

2014/11/15 ～ 2014/11/19にかけて，Walter E. Washington Convention Centerにて開催されましたNeuroscience2014に参加いたしました．本大会は，脳神経に関係する研究者が世界中から集まり，ポスターやオーラル，シンポジウムなどでそれぞれの研究内容について議論を深め，新しい知見を得ようとする非常に大きい学会で，Society for Neuroscienceによって主催されました．
私は全ての日程に参加いたしました．本研究室からは他に山本先生，大村さん，大西さん，眞島さん，大谷，滝が参加しました．
 

2. 研究発表
   • 発表概要

私は18日の午前のポスターセッション「Human Emotion: Behavioral and Neural Mechanisms」に参加いたしました．発表の形式はポスター発表で，ポスターを見に立ち寄ってくれた方々に対して，私の研究についての説明や議論を行いました．
今回の発表は，マルチタスク時の心理状態と脳活動の関係を検討したものです．発表タイトルは「Relationship between brain activity and emotional state during multi-task」です．以下に抄録を記載致します．

ObjectiveThe purpose of this study is to investigate the relationship between the brain activity and the changes in the emotional state during multi task. This paper describes that the emotional influence of multi-task differs from person to person even if they do the same task using simplified Profile of Mood States (POMS). We measure the brain activation during the multi task using functional Magnetic Resonance Imaging (fMRI), to find out a relationship between the brain activation and the emotional state.MethodsEight healthy young men and eight healthy young women (22±1 years old) participated in this study. The subjects performed online letter matching tasks, which provide a multi task and a single task. The multi task consists of a simultaneous memory of two problems at the same time. On the other hand, the single task offers that the subject memories only one problem. In both of the experiments, we investigated the relationship between the brain activation and the emotional state. In the first experiment using fMRI, we measured the brain activation during the multi tasks and the single tasks. In the second experiment using POMS, we measured the emotional changes during both of the tasks. On the basis of the emotional changes caused by each task, we classified subjects to the high-stress group or the low-stress one.The subjects with increase in emotional changes are defined as the high-stress group and those with decrease of emotional changes are defined as the low-stress group. We compared the brain activation between the high-stress group and the low-stress group. Results There were the high-stress subjects and the low-stress subjects in both tasks. The emotional influences caused by each task differed from person to person even if they do the same task. When the subjects performed the multi task, the difference of activation in the Vermis between the high-stress group and the low-stress one was observed. When the subjects performed the single task, the difference of activation in the Cerebellum between the high-stress group and the low-stress one was exhibited. Although there is not a unique region about the multi task, the results suggested that the difference in the emotional state is concerned with the brain activation in Vermis. Conclusions The emotional changes in the multi task and the single task differed from person to person. The difference in emotional changes was concerned with the brain activation even if they do the same task. This paper suggested that the emotional state during the multi task is closely related to the brain activation in Vermis.

 

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
 
・質問内容1
質問者の氏名を控え損ねてしまいました．こちらの質問は，POMSではなく，PANASやSTAI，Manikin scaleなどのアンケートの方が，ストレスを調べることにおいて良いのではないか，というものでした．これに対して，POMSは実際に抑うつなどの指標として使用されているので用いたが，それらのアンケートについても今後調査したいとお答えしました．
 
・質問内容2
質問者の氏名を控え損ねてしましました．こちらの質問はHigh Stress群の高成績群とLow Stress群の高成績群では別の性質を持っているのでは，というものでした．この質問に対して，今後検討してみたいとお答えしました．
 
・質問内容3
質問者の氏名を控え損ねてしましました．こちらの質問はタスクの内容についてでした．この質問に対して，タスクによりマルチタスク状態が作られること，それによるストレス状態を想定していることをお答えしました．
 

• 感想

学会では，ストレスに対して動物を使用して研究している方や，心理状態を中心に研究している方が意見を下さり，大変刺激になりました．英語で研究について議論するのはとても難しかったですが，まったく違う国で研究している方々とコミュニケーションをとれることがとてもうれしかったです．次回に向けてこれからも研究と英語の勉強に励もうと思いました．
 

3. 聴講

今回の講演会では，下記の5件の発表を聴講しました．
 

発表タイトル       ：　EFFECTS OF ACUTE STRESS ON RISKY MONETARY DECISION-MAKING著者                  ： Peter Sokol-Hessner, Candace M. Raio, Sarah P. Gottesman, Sandra F. Lackovic, Elizabeth A. Phelpsセッション名       ： Decision-Making and StressAbstruct            ：A substantial body of recent research has argued for the importance of affective factors in human decision-making. However, the precise ways in which affect interacts with decision-making is still unclear, in part because it can be notoriously difficult to dissociate the specific processes and mechanisms underlying both. A number of recent studies have manipulated acute stress, an affective response with well-characterized neurohormonal components, to examine its effect on cognitive function. The promise of such a specific manipulation can only be fully realized, however, if an equal level of specificity is applied to the choices under examination. Only by separating the valuation and decision processes that underlie choices is it possible to identify the relationships that do, and do not, exist between those processes and the dimension of affect being manipulated. In risky decision-making, in which participants choose between options that vary in the amount that may be probabilistically won or lost, findings of the effects of stress have been mixed. Some studies have found that stress leads to risk aversion, while others have found that it leads to risk seeking. A limitation of many of these studies is that they fail to dissociate the processes that drive decision-making in risky situations, including not only risk attitudes but loss aversion and overall noisiness or consistency. Here, participants performed a standard risky monetary decision making task on each of two days. This task allows the quantification and separation of three distinct choice processes: risk attitudes (feelings about chance), loss aversion (the relative weighting of losses to gains), and choice consistency. At the beginning of each day, participants were equally likely to undergo an acute stress manipulation (the cold pressor task, or CPT, in which participants submerge their arm up to their elbow in ice water for three minutes) or a control condition (with room temperature water). We were thus able to examine participants’ estimated parameter values on each day, and quantify the contribution of stress to each component of choice, within-subjects. Using this econometric model of participants’ decision behavior, we found stress selectively affected risk attitudes by reducing risk aversion for gains, and did not affect loss aversion or consistency. Only by computationally separating the processes that contribute to risky monetary choice, and combining them with a well-characterized manipulation of the acute stress response, were we able to address confounds present in prior studies, and quantify the effects of stress in risky decision-making.

この発表は，ヒトが意思決定をする際のストレス状態が，意思決定の結果に及ぼす影響を検討したものでした．具体的にはCPTというタスクによってストレス状態を引き起こし，その度合いをコルチゾールによって計測し，金銭が関係するタスクを行わせたときの結果と比較した実験です．
私の研究では，タスクによるストレスとそのタスク時の脳活動の関係を検討しているのですが，この研究では，タスクを行わせる前にストレス状態を調整しておき，その状態によってタスクに影響が出るというところが面白いと思いました．
 

発表タイトル       ： Effect of acute stress on working memory in male rats sexually motivated著者                  ：E. HERNANDEZ-ARTEAGA, M. ALMANZA-SEPÚLVEDA, M. HERNÁNDEZ-GONZÁLEZ, M. GUEVARA, H. BONILLA-JAIME, M. OLVERA CORTÉSセッション名       ： Executive function: Learning and memoryAbstruct            ： Stress is the biological mechanism through which the body attempts to regain homeostasis when affected by internal and external forces (stressors), based on behavioral and endocrine mechanisms. Stress can affect prefrontal cortex and thus affect executive functions, such as working memory. On the other hand, it is known that sexual behavior is an effective reward that enhances the acquisition and maintenance of working memory. Therefore, this study determines the effect of acute stress on visuospatial working memory (non-matching-to-sample working memory task using a T-maze) in sexually motivated male rats. 32 sexually experts male rats were trained in a T-maze using sexual interaction as a reward during a 4-day training period. According to their performance, the rats were divided into 2 groups: good-learners (n=12) and bad-learners (n=20). Finally, on the fifth day, 6 good-learners and 10 bad-learners were stressed by cold water immersion (CWI, 15°C) for 15 minutes before their execution in the T-maze while the remaining rats formed the control group. The results showed that acute stress by CWI improved working memory in both groups (good- and bad-learners), and decreased their sexual motivation to perform the task.

この発表では，急性ストレスと性的動機がワーキングメモリに及ぼす影響を，ラットを用いて研究したものだった．ワーキングメモリは私の実験のタスクでも機能していると考えられるため，その成績とストレスの度合いにも関係がある可能性が示唆されました．
 

発表タイトル       ： Mindfulness Based Stress Reduction correlates with frontal lobe changes during mood induction fMRI著者                  ： B. B. BRADEN, T. K. GLASPY, T. PIPE, K. STEINKE, C. T. ELROD, L. C. BAXTERセッション名       ：Human Emotion: Behavioral and Neural MechanismsAbstruct            ：Depression neural circuitry has been established in part through brain mapping during induction of a sad mood. Interventions such as medication and deep brain stimulation have been shown to modulate depression neural circuitry. We examined whether a behavioral intervention, Mindfulness Based Stress Reduction (MBSR), similarly modulates regional brain activity during emotional responsivity. Participants were 24 back pain patients, 13 of whom completed a 4 week MBSR class. The 11 Control participants were given MBSR reading material with no further instruction. All participants were evaluated for Beck Depression Inventory-II and Oswestry Low Back Pain and scanned on the same 3T Philips scanner a week before and after the MBSR training. During scanning, we used an fMRI mood induction paradigm that maps neural changes during transitions between sad and neutral moods. During the task, participants viewed alternating sets of sad (mourning people) or neutral (landscape) pictures for mood induction, signaling that they achieved the desired mood state (sad or neutral) with a button press. Individualized regressors linearly modeled the time periods during the attainment of sad and neutral moods. Parameters were TE/TR= 30/3000ms, FOV =24, 60×60 matrix, 40 contiguous slices 4mm thick collected in the axial plane. SPM8 was used for image processing and analyses. Paired t-tests indicated that both groups declined in depression scores (Controls; p=0.05, MBSR p=0.08 but only the MBSR group showed a significant decline in back pain symptoms (p<0.02; p=0.1 for Controls). At baseline, there were no group differences in fMRI responsivity. Post-intervention, the MBSR group showed greater BOLD signal in the middle anterior cingulate cortex (ACC; p=0.035, corrected), subgenual ACC (p=0.043, corrected), and left insula (p=0.038, corrected) during transitions from a sad to a neutral state, than the Control group. In summary, despite a mild improvement in depression in both groups, only the MBSR intervention improved back pain symptoms. FMRI results parallel these findings, showing greater frontal system engagement in regions important in awareness of internal states (insula) and cognitive awareness of emotional regulation (ACC). Further research is warranted to evaluate the neural correlates of MBSR as a behavioral intervention for distressed patient populations, including chronic back pain.

この発表では，うつ病の改善に向けたアプローチとして，薬理学的なアプローチではなく，行動ベースのマインドフルネスなアプローチが有用かどうかを検討したものでした．この研究では，うつ病の改善の指標としてmiddle anterior cingulate cortex（ACC）, subgenual ACC, and left insula などの脳活動を指標としていました．それらの部位はストレスとも関連しており，私も着目している部位であるので非常に興味深いものでした．
 

発表タイトル       ： A cognitive function of the default mode network in monkeys: Shifting of selective attention?著者                  ：N. S. CASPARI, R. VANDENBERGHE, W. VANDUFFELセッション名       ：Executive Function: Network ActivityAbstruct            ：The default mode network (DMN) is a set of brain regions activated during rest in human (Shulman et al. 1997) and monkey (Mantini et al., 2011). It is engaged during internally focused tasks including autobiographical memory retrieval, envisioning the future, and conceiving the perspectives of others. In addition the DMN is hypothesized to support a broad low-level focus of attention when monitoring the external world for unexpected events (Buckner et al. 2008). Defining an overarching function common to such widely different conditions, however, is implicitly difficult. An intriguing possibility is, that ‘shifting operations’ between series of internal thoughts, memories, and during passive, seemingly indifferent observation of the environment might be the glue across these conditions. If so, shifts in spatial attention should also engage the DMN. We tested this hypothesis using monkey fMRI during a covert selective attention task, previously used in humans (Molenberghs et al. 2007). Two pairs of shapes were peripherally presented and each pair contained a relevant and irrelevant shape. Monkeys fixated in the center and had to respond manually when the relevant stimulus dimmed. An event consisted of the replacement of the current stimulus pair by the other pair. In 1/3 of the trials this change between pairs elicited a spatial shift in attention as the relevant stimulus was replaced by an irrelevant one. An event-related analysis (N=3) revealed a high degree of overlap (>70%) in cortex posterior to the CS, between shift-related activations and the monkey DMN as defined in the consensus map of Mantini et al. (2011) (comparing rest vs. active task conditions in 15 expts). In contrast, shift-related activations anterior to the central sulcus, overlapped only to 8.15% with the monkey DMN, indicating possible functional subdivisions of the DMN. In the precentral sulcus and the ACC, shift- and DMN specific activations clustered adjacently. Sustained contralateral attention overlapped with the stimulus representations and activated an entirely different set of areas, except for portions of the ACC, the IPS, and area 12. Our data show that the posterior core of the DMN is clearly activated during shifting attention from one location to another, potentially mediated by frontal areas during top-down attentional control. It is therefore tempting to hypothesize that shifting operations in general, be it across memories, thoughts, and internally generated representations, are one important defining feature of the DMN. Cognitive shifting operations also break down in DMN-associated pathologies such as Alzheimer’s disease, autism and schizophrenia (Buckner et al. 2008).

この発表では，Default Mode Network の役割の1つに，回想状態や未来への広く浅い注意の状態など様々な状態の間の変異も含まれているという仮説をサルを用いて検証したものだった．注意の切り替えは様々なタスク中に行われていると考えられ，私の実験にも関係があると思われる．この研究で着目していたprecentral sulcus と ACCについて，自分の実験でも注目したい．
 

発表タイトル       ： Clustering coefficient during resting state predicts trait anxiety in healthy adults著者                  ：X. YOU, E. GORDON, M. NORR, M. STOLLSTORFF, C. J. VAIDYAセッション名       ：Human Emotion: Individual Differences and DisordersAbstruct            ：Anxiety can be defined as an unpleasant state of mental uneasiness or concern that is accompanied by physical discomfort. An anxious temperament (termed trait anxiety) is a risk factor for mood disorders and has been associated with individual differences in activation of medial prefrontal cortex (PFC) and amygdala, regions involved in emotional processing. Resting-state functional networks involving those regions have been associated with trait anxiety using region-of-interest methods. Whether voxel-wise data-driven methods reveal those same associations is not known. Here, we examined whether a voxel-wise graph theory metric of clustering coefficient is sensitive to individual differences in trait anxiety (measured by the State-Trait Anxiety Inventory (STAI). Clustering coefficient is the probability that the neighbors of the local node are also connected to each other, as a measure of local connectivity or “cliqueness” of a graph. High clustering is associated with robustness of a network, which is resilience against random network damage. We predicted that trait anxiety will correlate with clustering coefficient in medial PFC and amygdala. Eighty undergraduates (ages 18-22) completed the STAI and underwent fMRI at 3T for 5 min in a resting state with eyes closed. 152 fMRI images were slice-time- and motion-corrected, normalized, smoothed with Gaussian kernel with FWHM of 8mm in SPM8, and bandpass filtered (0.01-0.1Hz). The images were down-sampled to 6 mm voxel size for computational efficiency. Time course of each voxel was correlated to every other voxel and thresholded at p=.001 FDR corrected (r > .32). Clustering coefficient maps were computed on the graph using the brain connectivity toolbox (https: //sites.google.com/site/bctnet/measures/list). These maps were entered into a regression with STAI trait anxiety scores entered as the covariate of interest and mean framewise displacement as a covariate of no interest (p < .05 monte carlo corrected). As predicted, higher trait anxiety scores were associated with higher clustering coefficient in ventral and dorsal medial PFC (BA10) extending to anterior cingulate (BA 32), regions implicated in emotional processing and monitoring. Contrary to prediction, a positive correlation was also observed in bilateral parahippocampus/fusiform area and visual cortex (BA 17/18), regions involved in memory and visual processing. Together, more robust information processing – high local connectivity of the graph in these regions during a task-free state may suggest the higher vigilance/monitoring behavior that accompanies higher emotional reactivity in people with higher trait anxiety.

この発表では，Resting-Stateの脳活動のクラスタリング係数が，State-Trait Anxiety Inventory（STAI）のスコアと関係がみられたというものでした．STAIは対象者の不安の状態や不安に対する性質を得点化するアンケートであり，私が今回の学会で受けたアドバイスの1つでした．次回の実験から検討する予定なので，その脳活動との関連について調べているこの研究は非常に興味深かった．
学会参加報告書

報告者氏名	大西夏子
発表論文タイトル	知的生産性向上のための色温度環境の検討-心理状態と脳活動に及ぼす影響の影響-
発表論文英タイトル	Analysis of color temperature to improve intelligent productivity -The effect of psychological states and brain activity-
著者	大西夏子，山本詩子，三木光範，廣安知之
主催	Society for Neuroscience
講演会名	Society for Neuroscience2014
会場	Walter E. Washington Convention Center
開催日程	2014/11/15-2014/11/19

 
 

1. 講演会の詳細

2014/11/15から2014/11/19にかけて，Walter E. Washington Convention Centerにて開催されましたSociety for Neuroscience2014 1) に参加いたしました．このSociety for Neuroscience2014は，Society for Neuroscienceによって主催された研究会で，生体計測および脳機能イメージング装置を用いてNeuroscienceにおける議論を行い，あらゆる生物学的組織における研究の促進や新たな科学的知識の応用により，治療法の開発や脳神経学の理解を深めることを目的に開催されています．
私は15日から19日の会期全てに参加いたしました．本研究室からは他に山本先生，眞島，大村，滝，岡村，大谷が参加しました．
 

2. 研究発表
   • 発表概要

私は19日の午後のセッション「Functional Mechanisms of Attention II」に参加いたしました．発表の形式はポスター発表で，240分の講演時間内に60分の発表時間となっておりました．
今回の発表は，色温度環境が選択的注意時に及ぼす影響を心理状態および脳活動を基に検討したものです．発表タイトルは「Analysis of color temperature to improve intelligent productivity -The effect of psychological states and brain activity-」です．以下に抄録を記載致します．
 

【Purpose】The aim of this study is to derive appropriate light environment to get a better work for office worker. Prior research mentioned that color temperature of office light effected on psychological states like fatigue. Therefore we investigate the psychological states before the start of experiment and discuss the effect of color temperature on brain function and working efficiency.【Methods】Thirty-four healthy adult male (age, 21.5 ± 0.5 years) participated in this study. Before the start of experiment, they completed POMS (Profile of Mood States) that was used to evaluate the psychological states. After that, subjects were exposed to two types of lights (high color temperature at 7966 ± 65 K and low color temperature at 3226 ± 28 K) during the GO/NOGO task. In order to investigate brain activity during sustained attention, the subjects performed the GO/NOGO task. We investigated the GO/NOGO task performance and cerebral blood flow (CBF) change during sustained attention. The performance was evaluated on the basis of an error rate and reaction time (RT) of the GO/NOGO task. Reaction of the subject was counted as error when the reaction time of go trials exceeds 500 ms. In addition, fast 10% was calculated the average of reaction times among the 10% of the fastest reaction times in each subject. We measured CBF changes in the inferior frontal cortex (IFC) under sustained attention using fNIRS (functional near-infrared spectroscopy).【Results and Discussion】We classified subjects into two groups ; “Vigor” and “non-Vigor”. Any significant difference between Vigor and non-Vigor was not seen in RT (p>.05). However, fast 10% was faster in Vigor group than non-Vigor group under both of lights. In non-Vigor group, CBF decreased or unchanged under both of lights. On the other hand, in Vigor group, CBF significantly increased under Blue light (p<.05). These results suggest that Vigor group demonstrated a greater IFC activation and high score when exposed to high color temperature than non-Vigorous group.【Conclusion】This study demonstrated that the effect of color temperature differs from psychological states of the start of experiment. The results suggest that, when exposed to high color temperature, vigorous people performed fast responses and more activated in IFC.

 

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
 
 
・知的生産性は何で測っているのか？
Alina Shevorlkinさんから頂いた質問です．
視覚探索課題の課題成績をもとに評価していると回答致しました．課題成績は反応時間を用いており，反応時間が早いほど知的生産性が向上したと定義し、検討を進めたと説明しました．
 
・アイトラッキングを用いて視点追跡をしたらどうか．
東京電機大学の我妻伸彦先生から頂いたアドバイスです．
ターゲットが提示される場所によって反応時間が変わる可能性がある．また，視点追跡により光環境間での追跡の方略が分かる可能性がある、とのアドバイスを頂きました．さらに，一点を見つめる長さを測ることにより，眠気を測る指標にもなると教えて頂きました．
 
・実験時間帯の統制は行ったのか．
東京電機大学の我妻伸彦先生から頂いた質問です．
13:00-17:00で統制を行ったと回答致しました．朝は眠気の影響が大きく，夜は疲労の影響が大きく出るため，実験を行う時間帯は13:00-17:00が一番良いとのことでした．実験時間は被験者に与える影響が大きいため気をつけなければいけないというアドバイスを頂きました．
 
・課題に対する慣れの影響はどうなのか．被験者間で慣れの差は出なかったのか．
The city university of New YorkのDaniel Roblesさんから頂いた質問です．
被験者は予備実験として本実験前に1回課題の練習を行った後に実験を行ったため，慣れの程度は統制させていると考えて実験を行ったと回答致しました．また口頭のアンケート結果より，被験者間で得意・不得意の差の影響は出ている可能性もあることも回答致しました．
 
・被験者の視力は考慮したのか．
立命館大学の横田さとし先生から頂いた質問です．
被験者の視力は測定しておらず，課題の回答に支障が出ない視力であることの確認のみであると回答致しました。0.7～1.0の視力が実験に適しており，1.5などの視力が良すぎる人も被験者の対象から除くべきであるとのアドバイスを頂きました．また，視野角も測るべきであると教えて頂きました．
 
・照明の色の変化はどのくらいの時間で目が慣れてくるのか．
慶応大学大学院の土橋渡さんから頂いた質問です．
5分程度で人間の目は光の色の変化に慣れてくると回答致しました．実験室の設計や照明制御についてとても興味深いと言って頂けました．家具の色調や素材によっても照明環境が与える印象は変わるので，家具との調和も重要であるとのアドバイスを頂きました．
 
・カンデラは測っていないのか．
東京電機大学の我妻伸彦先生から頂いた質問です．
カンデラは計測しておらず，照明環境のパラメータは色温度をケルビン，照度をルクスの2種類を用いたと回答致しました．
 

• 感想
  ポスター聴講に来て下さった一人一人とたくさん議論をすることが出来て良かったです．視覚研究の専門家の方に，アイトラッキングや視力のことなど参考になる話をたくさんお聞きすることができ，とても勉強になりました．また，昨年聴講に来て下さった方が今年も来て下さり，嬉しかったです．研究分野が異なる研究者の方も多く来て下さり，違った視点からアドバイスを頂けて参考になりました．注意や記憶の脳機能を調査しているポスターを中心に聴講して，さまざまな課題や解析手法があることが分かり，とても参考になりました．

3. 聴講

今回の講演会では，下記の5件の発表を聴講しました．
 

発表タイトル      ：　Conscious level required for working memory and long-term memory functions: An anesthesia study in humans著者            ： M. Osaka, K.ENDO, T.IKEDA, T.MINAMOTO, A.NAKAE, S.HAGIWARA, Y.FUJINO, T.MASHIMOセッション名       ： Working Memory　I   Abstract            ：Anesthesia is known to impair large variety of cognitive functions, disturbing neural synchronization. However, there are few studies systematically investigating an effect of anesthetic drugs on those functions. Manipulating a level of sedation, the present study examined the effect of anesthesia on two main cognitive functions: working memory and long-term memory. Participants received either Propofol (n = 10) or Midazolam (n = 10) and their cognitive performance was measured in four different sedative phases: before anesthesia (baseline), deep, moderate, and light phases. The sedation phases were manipulated individually with a target control infusion. We measured a reference level that produced conscious loss based on an eyelash reflex, and decreased the level in 1/4 step for each phase: 3/4 level for the deep, 2/4 for the moderate, and 1/4 for the light. Working memory was measured with a dual task paradigm where participants were required to remember words in a target category while ignoring non-category words. Following the task, a recognition task for the target words was given in order to test long-term memory. Another recognition task was given after the experiment when conscious level was fully recovered. Five participants were removed due to extremely poor performance (3 SDs away from the mean). In the working memory task, performance in the deepest phase was lower in comparison to the baseline and light conditions across two drugs. The result indicates that more than half of the conscious level was required for normal working memory function. On the other hand, in the first recognition task, performance in the deep condition was lower than those in other three conditions, indicating that half of the conscious level allows for normal long-term memory. Interestingly, in the second recognition task, performance in the moderate sedation was significantly lower than those in the baseline and light sedation. Those results indicate that memory transference is required three-quarter of the consciousness.

この発表は，麻酔の投入による意識レベルの違いがワーキングメモリと長期記憶に及ぼす影響を検討したものでした． 意識レベルの違いで記憶に差が出るのがおもしろいと思いました．
麻酔を使った実験は初めて聞いたので，被験者への倫理的配慮が難しそうだと感じました．ワーキングメモリの専門家である苧阪先生のお話を直接お聞きでき，非常に貴重な体験をしました．
 

発表タイトル      ：Binding across long-term memory and short-term memory in amnesia著者                  ：Elizabeth Race, Keely Burke & Mieke Verfaelieセッション名       ：Human Long-Term Memory: Medial Temporal Lobe II Abstract            ：It is known that short-term memory (STM) can improve when to-be-remembered information appears in contexts that make contact with existing long-term memory (LTM) representations. Recently, we demonstrated that the medial temporal lobe (MTL) plays a critical role in such integration across STM and LTM. Specifically, amnesic patients with MTL damage could not leverage intact visuospatial LTM representations to support verbal STM (Race et al., 2014). An important outstanding question is whether the MTL also plays a critical role in within-domain binding across LTM and STM. To investigate this question, we tested whether MTL amnesics could leverage intact verbal LTM to support verbal STM performance. Patients and matched controls immediately recalled lists of words that were either presented in familiar verbal contexts (word lists appeared as meaningful sentences) or unfamiliar verbal contexts (random word lists). Both patients and controls demonstrated enhanced immediate recall of verbal material in familiar verbal contexts (sentence advantage effect), indicating that patients are able to use intact semantic LTM to benefit verbal STM. These results reveal that the MTL is not critical for integrating across STM and LTM when that integration occurs within a single domain, consistent with a prior report of intact within-domain STM-LTM binding in developmental amnesia (Baddeley et al., 2010).

この発表は，内側側頭葉が短期記憶と長期記憶を統合する際に重要であるという報告でした．内側側頭葉の損傷患者と健常者に対して，数字を見たままキーボードに打つ単純課題と単語を並び替えて文章を作成する文章課題で実験を行っていました．内側側頭葉損傷患者と健常者では記憶の量に大きな差異が出来ていることに驚きました．被験者が60歳代前半の方だったので，若年の被験者で同様の実験を行うと結果がどのようになるのか気になりました．
 
 

発表タイトル      ：Reward- and conflict- induced interference: guiding and biasing of visual attention著者                  ：C.M.Giattiono, B.van den Berg, F.B. Lee, M.M. Lorist, and M.G. Woldorffセッション名       ：Functional Mechanisms of Attention II. Abstract            ：Features in the world around us are constantly biasing our attention. When these features are associated with reward (and are thus more salient), they can bias our attention, thereby promoting – or sometimes impairing – cognitive task performance, depending on the relevance of those features for our goals. Here we examined how visual attention and conflict processing interact with reward associations. Participants performed a version of the Erikson flanker paradigm in which the central target letter could be surrounded by symmetric flankers (e.g., XXYXX [incongruent] or YYYYY [congruent]) or asymmetric ones (e.g., XXYYY, left congruent, right incongruent), while neurophysiological (EEG) data was recorded, from which we extracted time-locked event-related potential (ERP) and event-related spectral (ERSP) responses. Additionally, two of a set of four letters were associated with high reward and the other two with low reward, all of which could act either as targets or flankers in different trials. Participants could gain money depending on their response time (RT) performance. Results showed that RTs increased as the number of incongruent flankers increased (bilateral congruent RTs < asymmetric incongruent RTs < bilateral incongruent RTs), as well as when the target letter was associated with high reward compared to low reward. In addition, if flanking incongruent letters were associated with high (vs. low) reward, participants responded more slowly, suggesting a reward-induced increase in attentional distraction. Asymmetrical conflict effects were observed in the ERPs with a lateralized negative deflection (at ~250-400 ms), followed by an increase in oscillatory power in the alpha band (8-12 Hz) substantially later (at ~600-900 ms), suggesting an initial distraction-induced attentional orienting toward the conflicting letters, followed by an active suppression of that input. In addition, preliminary analyses indicated that bilateral high-reward flankers elicited enhanced attentional biasing, as reflected by a decrease in occipital alpha. Finally, the analyses suggested that high-reward targets (irrespective of flanker reward) increased attentional biasing activity frontocentrally, manifested by a decrease in oscillatory beta-band activity (12-18 Hz). Together, these results reveal the different mechanisms by which reward can guide attention to suppress the processing of irrelevant distracting features (e.g., incongruent flankers) and to enhance the processing of relevant features (e.g., reward-associated targets) via top-down biasing.

この発表は，EEGを用いた研究で，報酬の高い課題時にはα波のパワー値が著しく減少し，注意レベルの向上と課題の反応時間の増加がみられたことを報告したものです．脳波の増減をカラーマップで示しており，高報酬と低報酬の差が大きく表れていることが良くわかり，おもしろいなと思いました．脳波の左右差の検討手法もお聞きすることができ， とても勉強になりました．
 
 

発表タイトル      ：Cognitive difficulties in urban population of cigarrets smokers著者                  ：A. SHEVORYKIN, D. ROBLES, J. MOSES, R. D. MELARAセッション名       ：Functional Mechanisms of Attention II. Abstract            ：Understanding the social, psychological, and neurological mechanisms that drive addicts to crave and seek drugs of abuse is an essential first step in developing effective treatments for addictions. Nicotine dependence from tobacco smoking remains among the most common form of addiction in the United States and worldwide, with 250 million packs of cigarettes consumed each day and 5.4 million individuals die each year from smoking-related causes (CDC, 2012). Despite numerous behavioral and pharmacological techniques currently available to quit cigarette smoking, the relapse rate for nicotine addiction is among the highest of any addictive substance, with 75-80% of quitters relapsing within six months (Zhou, Nonnemaker, Sherrill, et al., 2009). The goal of this study was to examine the neural and behavioral correlates of tobacco craving in cigarette smokers during the performance of a cognitive task. A group of smokers and a group of nonsmokers (matched in age, education and gender) performed a version of the flanker conflict task (discriminate lines preceded and followed by matching or mismatching lines) in the presence of one of four visual cues (smoking, positive, negative, and neutral faces) while EEG was recorded. Smokers refrained from cigarettes for one hour prior to testing. The flanker effect – an index of the magnitude of distraction from conflict – was measured as the difference in reaction time on congruent (matching) versus incongruent (mismatching) trials. The flanker effect to neutral cues was comparable in smokers and nonsmokers. However, the flanker effect to smoking cues was significantly greater in smokers. The results suggest that tobacco-related visual cues are disruptive to normal processes of selective attention in cigarette smokers. Therefore, as potential mechanisms influencing these processing-related bias in smokers can be identified and analized, the substance abuse intervention approach could become more effective at targeting and reducing relapse rates among the smoking population. References Centers for Disease Control and Prevention. (2012). Smoking & Tobacco Use. Retrieved from http://www.cdc.gov/tobacco/ Zhou, X., Nonnemaker, J., Sherrill, B., Gilsenan, A.W., Coste,F., West, R. (2009). Attempts to quit smoking and relapse: Factors associated with success or failure from the ATTEMPT cohort study. Addictive Behavior, 34, 365-373

この発表は，喫煙者は非喫煙者と比較して選択的注意課題の反応が遅く，喫煙が選択的注意を行う際の脳内ネットワークに影響を及ぼす可能性があるという報告でした．喫煙によりストループ課題など認知課題を行う際に必要とされる脳のネットワークを破壊する可能性があることに驚きました．喫煙を含めた薬物などの中毒者や依存者の減少を目的とし，研究活動を行っているとのことでした．
 
 

発表タイトル      ：Parametric Manipulations in Simon and Go/NoGo reveal 　specificity of neural mechanisms of Response Selection and Inhibition著者                  ：S. WIJEAKUMAR, M. W. VOSS, V. A. MAGNOTTA, A. T. BUSS, R. E. HAZELTINE, J. P. SPENCERセッション名       ：Functional Mechanisms of Attention II. Abstract            ：Increased activation in cortical and sub-cortical regions has been reported in response to infrequent inhibition. However, it is unclear whether this increase in activation is due to infrequent inhibition or infrequent trials in general. Further, it is unknown how neural activation changes in response to increased stimulus-response mappings within the context of response selection. We used functional magnetic resonance imaging to investigate the effects on behavior and neural activation for two parametric manipulations in a Simon and a Go/NoGo (GnG): a working memory load manipulation implemented by varying the number of stimulus-response mappings and a proportion manipulation implemented by varying the proportion of excitatory trials. It has also been shown that functional networks involved in task performance have enhanced synchrony during the resting state directly following task engagement. Therefore, we compared change in resting state networks before to after each task to determine whether the tasks would elicit distinct network interactions. Seed-based analyses were used to extract the default-mode network (DMN), salience network (SALN), executive control network (ECN) and a sensorimotor network (SMN). Behavioral results revealed increased reaction times on infrequent trials and also an increase in reaction times with an increasing number of stimulus-response mappings (load) for both tasks. A broad network of cortical, sub-cortical and cerebellar areas were activated by both tasks. Activation was greater on Go/compatible than Nogo/Incompatible trials at areas involved in motor-planning and control of movement. Bilateral lingual gyrus was selectively activated and de-activated in the Simon and GnG tasks respectively, reflecting a role specific to visuo-spatial attention. Insular-thalamic regions were selectively activated on infrequent events across both tasks. With regard to the load manipulation, there was a decrease in activation in the right inferior parietal lobule at higher loads. Thus, in contrast to the typical increase in neural activation with greater working memory demands, we found a decrease in activation suggesting that an associative memory mechanism underlies the stimulus-response mappings. Reliable resting state networks were identified across all resting-state runs. Both the SALN and ECN showed changes in resting state synchrony following both Response selection tasks. On the other hand, change in synchrony of the posterior cingulate cortex with the DMN was observed following the GnG task. Future work will integrate current theories and models of response selection with changes in neural activation.

この発表は，GO/NOGO taskでより難易度が高い課題時に視床領域の活性化と右下頭頂小葉での活性の減少がみられたという報告でした． 反応抑制や注意を司る背外側前頭前野や下前頭回以外での脳活動を検討していたため，とても興味深かった発表でした．今後の展望で選択的反応抑制の脳活動モデルを作成していくとのことだったので，モデル作成についても議論できてとても勉強になりました．
 
参考文献

• Society for Neuroscience2014，http://www.sfn.org/annual-meeting/neuroscience-2014

学会参加報告書

報告者氏名	滝謙一
発表論文タイトル	学会参加報告報告書
発表論文英タイトル	Analysis of processing mechanism of the human brain with multiple sensory information using fNIRS
著者	滝謙一, 山本詩子, 廣安知之
主催	Society for Neuroscience
講演会名	Neuroscience2014
会場	Walter E. Washington Convention Center
開催日程	2012/04/17-2012/04/20

 
 

1. 講演会の詳細

2014/11/15から2012/04/19にかけて，Walter E. Washington Convention Centerにて開催されましたNeuroscience2014に参加いたしました．このNeuroscience2013は，Society for Neuroscienceによって主催された研究会で，生体計測および脳機能イメージング装置を用いてNeuroscienceにおける議論を行い，あらゆる生物学的組織における研究の促進や新たな科学的知識の応用により，治療方法の開発や脳神経学の理解を深めることを目的に開催されています．
私は15から19日の間参加いたしました．本研究室からは他に山本先生，眞島さん，大西さん，大村さん，大谷さん，岡村さんが参加しました．
 

2. 研究発表
   • 発表概要

私は17日の午前のセッション「Perception and Imagery」に参加いたしました．発表の形式はポスター発表で，60分の講演時間となっておりました．
今回の発表は，視覚刺激と聴覚刺激を単一および同時に提示した際の脳活動の違いを検討したものです．以下に抄録を記載致します．

【Purpose】The brain treats sensory information and grasps surrounding circumstances. In a daily life, the brain processes not only single sensory information but also multiple sensory information. This leads to perform an accurate perception. From this fact, it is assumed that brain activities are different between single and multiple sensory stimulation. In this study, brain activities of single and multiple sensory stimulation are measured by functional Near-Infrared Spectroscopy (fNIRS) and a processing mechanism of sensory information is discussed.   【Method】 In the experiment, fNIRS device (ETG-7100 : Hitachi Medical Co., Ltd.) was used. The whole brain was measured with 116 channels. The subjects were 15 (10 male) healthy adults, ages 22-24. In the experiment, three types of stimuli were prepared; visual stimuli, auditory stimuli and audio-visual stimuli. In the visual stimuli, checkerboard which was reversed at 7.5Hz was presented. In the auditory stimuli, sine-wave sound of 1000Hz was presented. In the block design, there were the rest for 30s and the task for 15s. This lock design was repeated for five times. We performed three trials for each stimulus. During the task, subjects continue to look at the center of screen where the fixation point was marked. The derived cerebral blood flow data was processed with arithmetic mean the activation was discussed.   【Results】 The area that was active only when the audio-visual stimulus was presented was existed around both of temporal poles and the area that was active only when the single stimulus was presented was fund in the parietal region. In the other research, it was reported that the parietal association area involves the recognition of distance from stimulus. Thus, the subject may measure the distance between the subject and the stimulus presentation location and the parietal cortex was active. When audio-visual stimulus was presented, the parietal cortex was not activated. This result suggests that the brain activities prioritized to understand the relationship of the audio-visual stimulus compared to understand the positional relationship between the stimuli.   【Conclusions】 The brain activities of single and multiple sensory stimuli were examined. In the audio-visual stimulus presentation, the area around both the temporal poles was active. The parietal cortex was active when the single stimulus was presented. It is suggested that brain gives priority to understand the audio-visual stimulus relationship rather than the measurement of the positional relationship between the stimulus and subject.

 

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
また質問者の所属を控えるのを失念しておりました．
・活性CHの色付けの意味について
こちらの質問は活性CHの色が何を意味するかというものでした．この質問に対し私はそれぞれどの刺激で活性したかを示していると回答しました．
 
・何故fNIRSを使用したか
この質問に対しては音を聞かせる実験を行いたかったため騒音の大きいMRIは不適切であると判断したためと回答しました．
 
・何故ブロックデザインの1と5ブロック目のデータを除外したのか
1ブロック目は不慣れなためか血流が安定せず5ブロック目は疲労の影響があると判断したためと回答したつもりでしたがうまく伝えられませんでした．そのため外部に出している論文はあるかと訊ねられましたが英語で書いたものはないと答えました．
 
・今回の提示刺激では統合は起きないのではないか
統合部位の活性が報告されている実験設計を参考に新たに計測を行っていると回答いたしました．なお提示刺激も紹介したのですが音声のラウドネスや複雑さが統一されていない可能性についてご指摘を頂きました．
 
・pSTSの活動はfNIRSで計測可能なのか
この質問に対しては明確な回答をすることができませんでした．
 
・他のクロスモーダルの研究は行わないのか
現時点では視聴覚の統合のみを研究するつもりだと回答しました．
 
・pSTSの活動を捉えて今後どのような検討を行っていくのか
この質問に対しては，コネクティビティを検討しpSTSが感覚野から入力を受けているか確認するつもりですと回答しました．
 

• 感想
  3月の学会で発表した内容での発表だったため現在行っている実験についての話があまりできなかったことが残念です．また学会に参加する際はその締め切りを意識して報告できる内容を揃えておこうと思います．しかし初の国際学会を無事にこなせたことで自信につながりました．参加してよかったです．さらに深い議論を行えるように英語の学習をして次の国際学会に臨みたいと思います．

3. 聴講

今回の講演会では，下記の5件の発表を聴講しました．
 

発表タイトル       ：　Multisensory inputs enhance neural target discrimination in monkey prefrontal cortex著者                  ： F. KATSUKI, M. SAITO, M. A. BURT, T. R. STANFORD, B. A. ROWLAND, B. セッション名       ： Multisensory and Temporal Factors in Cross-Modal Processing Abstruct ： Behavioral performance can be enhanced by integrating cross-modal signals. This hasbeen well-characterized in cue detection/localization tasks, where spatially aligned cross-modalstimuli elicit faster and more accurate responses. These behavioral enhancements are correlatedwith response enhancements in superior colliculus neurons. Such enhancements are dependenton the spatial alignment of the cross-modal cues, which may reflect a superordinate rule ofmultisensory integration or the informational gains conferred in the specific context oflocalization. We explored this issue for more complex decisions, requiring detection of a visualtarget of varying difficulty (oddball color stimulus) and release of a lever when it was present.We evaluated the impact of cross-modal cues on behavioral decisions and responses of monkeydorsolateral prefrontal cortex (dlPFC) neurons. Neuronal responses were evaluated based ontarget discrimination time by comparing the firing rate between target and non-target stimuli. Inexperiment 1, monkeys were trained on an explicit visual discrimination task (auditory stimulirequired a no-go response), and tested with visual, auditory, and spatiotemporally concordantvisual-auditory stimuli, where the auditory stimulus provided no task-relevant information aboutwhich stimulus was the oddball. There were no significant behavioral or physiologicalenhancements (66 dlPFC neurons were studied) in the cross-modal versus the visual-alonecondition in this context. In experiment 2, different auditory tones (high versus low frequency)were paired with specific visual stimuli (oddball target present or absent) requiring either a go orno-go response to make them task-informative. Cross-modal stimuli now yielded more accuratebehavioral responses. Visually-responsive dlPFC neurons demonstrated faster discrimination forthe cross-modal than the visual component stimulus (visual: 112 versus 149 ms Permutation test,p<0.05), even in neurons not explicitly responsive to auditory stimuli (n=27). The resultsdemonstrate that behavioral and physiological multisensory enhancements were not contingenton the spatial proximity of the cross-modal cues when the auditory stimulus was not directlylinked to the decision. Rather, multisensory benefits were evident only when both modalityspecificcues provided task-relevant information about the presence of the target.

この研究はサルを用いて感覚間相互作用を検討しており上丘を多感覚統合領域として着目していました．私はfNIRSを用いた研究をしていることもあり大脳皮質以外の多感覚統合領域のことを知りませんでした．この発表を聞くことで上丘の存在と現在の知見を得られたのは収穫でした．
 

発表タイトル       ：fNIRS and fMRI signals are concordant during a bipedal motor task著者                  ： J. A. NOAH, Y. ONO, Y. NOMOTO, S. SHIMADA, A. TACHIBANA, X. ZHANG, S. BRONNER, J. HIRSCH セッション名       ： Multisensory: Cross-Modal Processing in Humans Abstruct            ： The academic success of students is a major concern in American culture; thus, thereis constant pressure for new discoveries that can improve education. Many classrooms arebeginning to substitute standard chairs with therapy balls, which help to improve students’ focusand classroom performance, according to teacher and student reports. Therapy balls are a type ofheightened sensory tool, which are often used in physical and occupational therapy as a strategyfor individuals with learning or sensory differences. Heightened sensory tools increase thesensory information that the brain receives and have been effective at improving attention andclassroom performance. However, no studies explain why these tools are successful. Anobservational study indicated that attention and academic task improved with the use of therapyballs compared to standard chairs and that the effect of vestibular and proprioceptive inputshould be further investigated in a laboratory setting. Researchers performed an experiment withelementary school age participants (N = 20) to examine the effect of heightened sensorystimulation on the performance of functional school tasks and standard balance tasks. Subjectsperformed math and comprehensive reading tests during seating on a standard chair, seating withincreased vestibular input, and seating with increased proprioceptive input. They also completedstatic balance tasks with eyes open/closed on a firm/foam surface using the Biodex Balancesystem. Preliminary results suggest that with the utilization of therapy balls, school functioneither is the same or better compared to a standard chair. In addition, balance measures improvewhen proprioceptive and visual input are occluded or distorted after spending time on thestability ball. This research has the potential to help develop specific and evidence based trainingfor teachers and students on the appropriate strategies to use alternative seating in a classroomsetting.

この研究はfNIRSとfMRIで得られる信号の相関性を検討するもので，両機器で得られる信号に大きな差があると感じていた私としてはすごく興味のある内容でした．結果としては多感覚統合領域において大きな相関値が得られたとありますが，どのような解析をしたかは聞き取れなかったため頂いたポスターを読み込んでいこうと思います．
 

発表タイトル       ：The effect of difficulty of motor actions on mirror neuron system activity during imitation learning著者                  ： T. I. HIMBERG, L. HIRVENKARI, R. K. HARI セッション名       ： Human Social Cognition: Behavior and Pharmacology Abstruct            ： Mirror neuron system (MNS) is considered to play an important role in understandingand imitation learning of other’s action. Using near-infrared spectroscopy (NIRS) weinvestigated whether the MNS activity during observation of sequential actions is related to theperformance of the subsequent imitation. The subjects observed movie stimuli in which themodel folded an origami (a piece of colored paper). We prepared two different action sequencesto fold the origami (easy and difficult conditions). The subject was instructed to watch the moviestimuli (observation phase) and later fold the origami with the same sequence as the videodemonstration (execution phase). For the behavioral analysis, we scored the imitationperformance by assessing the recorded video. We added one point per step if the subject foldedthe origami in the same procedure as the demonstration. There was a significant difference inbehavioral score between the easy and difficult conditions: the score for the easy condition wassignificantly greater than that for the difficult condition (t(15)=5.27, P<0.05). For NIRS dataanalyses, we applied 2 (conditions)×2 (phase) 2-way ANOVA. The result showed that there wasa significant interaction between those factors in the bilateral supramarginal gyrus (ch16 and 40,P <0.05). Subsequent analyses showed that those channels showed greater activity in the easycondition than in the difficult condition during the observation phase, while the opposite resultwas obtained in the execution phase. We then investigated the effective connectivity of thefronto-parietal cortices during imitation learning by means of psychophysiological interactions(PPI). The result showed that the left premotor cortex and the left supramarginal gyrus werestrongly connected in the difficult conditions compared to the easy conditions (P <0.05).Suggesting that broader fronto-parietal MNS regions were recruited in the difficult condition.These findings suggest that MNS is involved in imitation learning, and its activity andconnectivity among these areas vary with the degree of difficulty of the observed action.

明治大学の学生で同じくfNIRSを用いて研究している方で解析について貴重な意見を頂きました．具体的にはt検定による賦活部位の判定の際の多重比較補正の必要性，コネクティビティを見る手法にpsychophysiological interactionsがある，などです．
 
 

発表タイトル       ：Common brain activity patterns during perception, imagery, and dreaming著者                  ： T. HORIKAWA, Y. HOSOKAWA, Y. KAMITANI セッション名       ： Human Cognition and Behavior Abstruct            ：Dreaming often accompanies visual experience, which appears to be phenomenally similar to stimulus-induced perception, and also to top-down mental imagery. It has been debated whether the neural mechanism underlying dreaming is perception-like or imagery-like. Our previous study (Horikawa et al., 2013) has shown that visual dream contents (object categories) can be predicted from fMRI activity patterns in the visual cortex during sleep by machine-learning decoders trained on brain activity induced by stimulus images containing objects. This “perception-to-dream” decoding indicates that perception and dreaming share neural representations of visual contents. Here we extend this approach to investigate the commonality and difference of neural representations between perception, imagery, and dreaming in brain areas known to be responsive to perception and/or imagery tasks. In addition to decoders trained on stimulus-induced brain activity (perception-trained decoders), we used decoders trained on brain activity during a task in which subjects visually imagined objects (imagery-trained decoders), to predict dreamed objects from brain activity during sleep. We found that the overall accuracies for the decoding of dreamed objects were comparable between the perception-trained decoders and the imagery-trained decoders. Interestingly, the perceptiontrained decoders outperformed the imagery-trained decoders in relatively lower-level areas, the ventral visual cortex and inferior parietal cortex, while the imagery-trained decoders outperformed the perception-trained decoders in higher-level areas, the dorsolateral prefrontal cortex and hippocampus. These results suggest that dreaming shares neural representations with perception and imagery in multiple brain areas with greater similarity to perception and imagery in lower and higher areas, respectively. Thus, the neural representation of dreaming may not be unilaterally perception-like or imagery-like, but consist of a unique mixture of perception-like and imagery-like brain states.

これは夢を見ているときの脳活動と覚醒時に物体を見ているときの脳活動の類似性を検討している研究です．解析には機械学習を用いており，複雑な処理を行う脳機能の解明と機械学習の相性の良さを感じた．
 

発表タイトル       ：The role of multisensory signals in interval timing: An MEG study著者                  ： T. W. KONONOWICZ, L. LECOUTRE, V. VAN WASSENHOVEセッション名           ：Human Cognition: Timing and Temporal Processing Abstruct            ：Duration perception has been shown to be influenced by the sensory modality; for instance, for the same physical durations, auditory stimuli have often been reported to be perceived as longer and more accurately than visual stimuli (Penney, Gibbon, and Meck, 2000). This auditory dominance is even present when audio-visual durations are used (Burr, Banks, Morrone, 2009; Chen and Yeh, 2009). In the context of internal clock which integrates temporal pulses to provide an estimates of elapsed time, auditory dominance is believed to result from the enhanced pacemaker rate, something that could be reflected in a build up of sustained magnetic fields (Sieroka et al., 2003) . Alternatively, duration and accuracy of temporal performance could be reflected in a pre- and pos- interval neural dynamics (Mayo and Sommer, 2013). Given that the neural signatures of multisensory duration perception have been rarely investigated, we asked participants to discriminate subsecond intervals marked by Audio, Visual, and Audio-Visual stimuli. This study was also designed to address how expectancy of a certain modality can modulate duration perception and how this expectancy changes pre-interval dynamics. To this end, participants were cued with cues that were predictive of interval modality. We will demonstrate how interval modality and cue predicting interval modality impacts subjective timing. Moreover, the role of ramping activity (Wittman, 2013) and oscillatory dynamics in duration discrimination will be discussed (Kosem, Gramfort, and Van Wassenhove, 2014).

この研究は感覚刺激の知覚時間が特定の感覚刺激の提示感覚の予測によってどのように変化するかを調べた物です．異なる感覚刺激が同じ現象であるかどうか判断する条件としてどれだけの時間間隔でそれらが生じているかという研究は有名ですが，これはその後の統合について研究しており，多感覚統合の例には様々な物があり面白いと思いました．
 
参考文献

• Neuroscience2014，http://www.sfn.org/annual-meeting/neuroscience-2014

 
学会参加報告書

報告者氏名	滝謙一
発表論文タイトル	学会参加報告報告書
発表論文英タイトル	Analysis of processing mechanism of the human brain with multiple sensory information using fNIRS
著者	滝謙一, 山本詩子, 廣安知之
主催	Society for Neuroscience
講演会名	Neuroscience2014
会場	Walter E. Washington Convention Center
開催日程	2012/04/17-2012/04/20

 
 

1. 講演会の詳細

2014/11/15から2012/04/19にかけて，Walter E. Washington Convention Centerにて開催されましたNeuroscience2014に参加いたしました．このNeuroscience2013は，Society for Neuroscienceによって主催された研究会で，生体計測および脳機能イメージング装置を用いてNeuroscienceにおける議論を行い，あらゆる生物学的組織における研究の促進や新たな科学的知識の応用により，治療方法の開発や脳神経学の理解を深めることを目的に開催されています．
私は15から19日の間参加いたしました．本研究室からは他に山本先生，眞島さん，大西さん，大村さん，大谷さん，岡村さんが参加しました．
 

2. 研究発表
   • 発表概要

私は17日の午前のセッション「Perception and Imagery」に参加いたしました．発表の形式はポスター発表で，60分の講演時間となっておりました．
今回の発表は，視覚刺激と聴覚刺激を単一および同時に提示した際の脳活動の違いを検討したものです．以下に抄録を記載致します．

【Purpose】 The brain treats sensory information and grasps surrounding circumstances. In a daily life, the brain processes not only single sensory information but also multiple sensory information. This leads to perform an accurate perception. From this fact, it is assumed that brain activities are different between single and multiple sensory stimulation. In this study, brain activities of single and multiple sensory stimulation are measured by functional Near-Infrared Spectroscopy (fNIRS) and a processing mechanism of sensory information is discussed.   【Method】 In the experiment, fNIRS device (ETG-7100 : Hitachi Medical Co., Ltd.) was used. The whole brain was measured with 116 channels. The subjects were 15 (10 male) healthy adults, ages 22-24. In the experiment, three types of stimuli were prepared; visual stimuli, auditory stimuli and audio-visual stimuli. In the visual stimuli, checkerboard which was reversed at 7.5Hz was presented. In the auditory stimuli, sine-wave sound of 1000Hz was presented. In the block design, there were the rest for 30s and the task for 15s. This lock design was repeated for five times. We performed three trials for each stimulus. During the task, subjects continue to look at the center of screen where the fixation point was marked. The derived cerebral blood flow data was processed with arithmetic mean the activation was discussed.   【Results】 The area that was active only when the audio-visual stimulus was presented was existed around both of temporal poles and the area that was active only when the single stimulus was presented was fund in the parietal region. In the other research, it was reported that the parietal association area involves the recognition of distance from stimulus. Thus, the subject may measure the distance between the subject and the stimulus presentation location and the parietal cortex was active. When audio-visual stimulus was presented, the parietal cortex was not activated. This result suggests that the brain activities prioritized to understand the relationship of the audio-visual stimulus compared to understand the positional relationship between the stimuli.   【Conclusions】 The brain activities of single and multiple sensory stimuli were examined. In the audio-visual stimulus presentation, the area around both the temporal poles was active. The parietal cortex was active when the single stimulus was presented. It is suggested that brain gives priority to understand the audio-visual stimulus relationship rather than the measurement of the positional relationship between the stimulus and subject.

 

• 質疑応答

今回の講演発表では，以下のような質疑を受けました．
また質問者の所属を控えるのを失念しておりました．
・活性CHの色付けの意味について
こちらの質問は活性CHの色が何を意味するかというものでした．この質問に対し私はそれぞれどの刺激で活性したかを示していると回答しました．
 
・何故fNIRSを使用したか
この質問に対しては音を聞かせる実験を行いたかったため騒音の大きいMRIは不適切であると判断したためと回答しました．
 
・何故ブロックデザインの1と5ブロック目のデータを除外したのか
1ブロック目は不慣れなためか血流が安定せず5ブロック目は疲労の影響があると判断したためと回答したつもりでしたがうまく伝えられませんでした．そのため外部に出している論文はあるかと訊ねられましたが英語で書いたものはないと答えました．
 
・今回の提示刺激では統合は起きないのではないか
統合部位の活性が報告されている実験設計を参考に新たに計測を行っていると回答いたしました．なお提示刺激も紹介したのですが音声のラウドネスや複雑さが統一されていない可能性についてご指摘を頂きました．
 
・pSTSの活動はfNIRSで計測可能なのか
この質問に対しては明確な回答をすることができませんでした．
 
・他のクロスモーダルの研究は行わないのか
現時点では視聴覚の統合のみを研究するつもりだと回答しました．
 
・pSTSの活動を捉えて今後どのような検討を行っていくのか
この質問に対しては，コネクティビティを検討しpSTSが感覚野から入力を受けているか確認するつもりですと回答しました．
 

• 感想
  3月の学会で発表した内容での発表だったため現在行っている実験についての話があまりできなかったことが残念です．また学会に参加する際はその締め切りを意識して報告できる内容を揃えておこうと思います．しかし初の国際学会を無事にこなせたことで自信につながりました．参加してよかったです．さらに深い議論を行えるように英語の学習をして次の国際学会に臨みたいと思います．

3. 聴講

今回の講演会では，下記の5件の発表を聴講しました．
 

発表タイトル       ：　Multisensory inputs enhance neural target discrimination in monkey prefrontal cortex 著者                  ： F. KATSUKI, M. SAITO, M. A. BURT, T. R. STANFORD, B. A. ROWLAND, B. セッション名       ： Multisensory and Temporal Factors in Cross-Modal Processing Abstruct ： Behavioral performance can be enhanced by integrating cross-modal signals. This hasbeen well-characterized in cue detection/localization tasks, where spatially aligned cross-modalstimuli elicit faster and more accurate responses. These behavioral enhancements are correlatedwith response enhancements in superior colliculus neurons. Such enhancements are dependenton the spatial alignment of the cross-modal cues, which may reflect a superordinate rule ofmultisensory integration or the informational gains conferred in the specific context oflocalization. We explored this issue for more complex decisions, requiring detection of a visualtarget of varying difficulty (oddball color stimulus) and release of a lever when it was present.We evaluated the impact of cross-modal cues on behavioral decisions and responses of monkeydorsolateral prefrontal cortex (dlPFC) neurons. Neuronal responses were evaluated based ontarget discrimination time by comparing the firing rate between target and non-target stimuli. Inexperiment 1, monkeys were trained on an explicit visual discrimination task (auditory stimulirequired a no-go response), and tested with visual, auditory, and spatiotemporally concordantvisual-auditory stimuli, where the auditory stimulus provided no task-relevant information aboutwhich stimulus was the oddball. There were no significant behavioral or physiologicalenhancements (66 dlPFC neurons were studied) in the cross-modal versus the visual-alonecondition in this context. In experiment 2, different auditory tones (high versus low frequency)were paired with specific visual stimuli (oddball target present or absent) requiring either a go orno-go response to make them task-informative. Cross-modal stimuli now yielded more accuratebehavioral responses. Visually-responsive dlPFC neurons demonstrated faster discrimination forthe cross-modal than the visual component stimulus (visual: 112 versus 149 ms Permutation test,p<0.05), even in neurons not explicitly responsive to auditory stimuli (n=27). The resultsdemonstrate that behavioral and physiological multisensory enhancements were not contingenton the spatial proximity of the cross-modal cues when the auditory stimulus was not directlylinked to the decision. Rather, multisensory benefits were evident only when both modalityspecificcues provided task-relevant information about the presence of the target.

この研究はサルを用いて感覚間相互作用を検討しており上丘を多感覚統合領域として着目していました．私はfNIRSを用いた研究をしていることもあり大脳皮質以外の多感覚統合領域のことを知りませんでした．この発表を聞くことで上丘の存在と現在の知見を得られたのは収穫でした．
 

発表タイトル       ：fNIRS and fMRI signals are concordant during a bipedal motor task 著者                  ： J. A. NOAH, Y. ONO, Y. NOMOTO, S. SHIMADA, A. TACHIBANA, X. ZHANG, S. BRONNER, J. HIRSCH セッション名       ： Multisensory: Cross-Modal Processing in Humans Abstruct            ： The academic success of students is a major concern in American culture; thus, thereis constant pressure for new discoveries that can improve education. Many classrooms arebeginning to substitute standard chairs with therapy balls, which help to improve students’ focusand classroom performance, according to teacher and student reports. Therapy balls are a type ofheightened sensory tool, which are often used in physical and occupational therapy as a strategyfor individuals with learning or sensory differences. Heightened sensory tools increase thesensory information that the brain receives and have been effective at improving attention andclassroom performance. However, no studies explain why these tools are successful. Anobservational study indicated that attention and academic task improved with the use of therapyballs compared to standard chairs and that the effect of vestibular and proprioceptive inputshould be further investigated in a laboratory setting. Researchers performed an experiment withelementary school age participants (N = 20) to examine the effect of heightened sensorystimulation on the performance of functional school tasks and standard balance tasks. Subjectsperformed math and comprehensive reading tests during seating on a standard chair, seating withincreased vestibular input, and seating with increased proprioceptive input. They also completedstatic balance tasks with eyes open/closed on a firm/foam surface using the Biodex Balancesystem. Preliminary results suggest that with the utilization of therapy balls, school functioneither is the same or better compared to a standard chair. In addition, balance measures improvewhen proprioceptive and visual input are occluded or distorted after spending time on thestability ball. This research has the potential to help develop specific and evidence based trainingfor teachers and students on the appropriate strategies to use alternative seating in a classroomsetting.

この研究はfNIRSとfMRIで得られる信号の相関性を検討するもので，両機器で得られる信号に大きな差があると感じていた私としてはすごく興味のある内容でした．結果としては多感覚統合領域において大きな相関値が得られたとありますが，どのような解析をしたかは聞き取れなかったため頂いたポスターを読み込んでいこうと思います．
 

発表タイトル       ：The effect of difficulty of motor actions on mirror neuron system activity during imitation learning 著者                  ： T. I. HIMBERG, L. HIRVENKARI, R. K. HARI セッション名       ： Human Social Cognition: Behavior and Pharmacology Abstruct            ： Mirror neuron system (MNS) is considered to play an important role in understandingand imitation learning of other’s action. Using near-infrared spectroscopy (NIRS) weinvestigated whether the MNS activity during observation of sequential actions is related to theperformance of the subsequent imitation. The subjects observed movie stimuli in which themodel folded an origami (a piece of colored paper). We prepared two different action sequencesto fold the origami (easy and difficult conditions). The subject was instructed to watch the moviestimuli (observation phase) and later fold the origami with the same sequence as the videodemonstration (execution phase). For the behavioral analysis, we scored the imitationperformance by assessing the recorded video. We added one point per step if the subject foldedthe origami in the same procedure as the demonstration. There was a significant difference inbehavioral score between the easy and difficult conditions: the score for the easy condition wassignificantly greater than that for the difficult condition (t(15)=5.27, P<0.05). For NIRS dataanalyses, we applied 2 (conditions)×2 (phase) 2-way ANOVA. The result showed that there wasa significant interaction between those factors in the bilateral supramarginal gyrus (ch16 and 40,P <0.05). Subsequent analyses showed that those channels showed greater activity in the easycondition than in the difficult condition during the observation phase, while the opposite resultwas obtained in the execution phase. We then investigated the effective connectivity of thefronto-parietal cortices during imitation learning by means of psychophysiological interactions(PPI). The result showed that the left premotor cortex and the left supramarginal gyrus werestrongly connected in the difficult conditions compared to the easy conditions (P <0.05).Suggesting that broader fronto-parietal MNS regions were recruited in the difficult condition.These findings suggest that MNS is involved in imitation learning, and its activity andconnectivity among these areas vary with the degree of difficulty of the observed action.

明治大学の学生で同じくfNIRSを用いて研究している方で解析について貴重な意見を頂きました．具体的にはt検定による賦活部位の判定の際の多重比較補正の必要性，コネクティビティを見る手法にpsychophysiological interactionsがある，などです．
 
 

発表タイトル       ：Common brain activity patterns during perception, imagery, and dreaming 著者                  ： T. HORIKAWA, Y. HOSOKAWA, Y. KAMITANI セッション名       ： Human Cognition and Behavior Abstruct            ：Dreaming often accompanies visual experience, which appears to be phenomenally similar to stimulus-induced perception, and also to top-down mental imagery. It has been debated whether the neural mechanism underlying dreaming is perception-like or imagery-like. Our previous study (Horikawa et al., 2013) has shown that visual dream contents (object categories) can be predicted from fMRI activity patterns in the visual cortex during sleep by machine-learning decoders trained on brain activity induced by stimulus images containing objects. This “perception-to-dream” decoding indicates that perception and dreaming share neural representations of visual contents. Here we extend this approach to investigate the commonality and difference of neural representations between perception, imagery, and dreaming in brain areas known to be responsive to perception and/or imagery tasks. In addition to decoders trained on stimulus-induced brain activity (perception-trained decoders), we used decoders trained on brain activity during a task in which subjects visually imagined objects (imagery-trained decoders), to predict dreamed objects from brain activity during sleep. We found that the overall accuracies for the decoding of dreamed objects were comparable between the perception-trained decoders and the imagery-trained decoders. Interestingly, the perceptiontrained decoders outperformed the imagery-trained decoders in relatively lower-level areas, the ventral visual cortex and inferior parietal cortex, while the imagery-trained decoders outperformed the perception-trained decoders in higher-level areas, the dorsolateral prefrontal cortex and hippocampus. These results suggest that dreaming shares neural representations with perception and imagery in multiple brain areas with greater similarity to perception and imagery in lower and higher areas, respectively. Thus, the neural representation of dreaming may not be unilaterally perception-like or imagery-like, but consist of a unique mixture of perception-like and imagery-like brain states.

これは夢を見ているときの脳活動と覚醒時に物体を見ているときの脳活動の類似性を検討している研究です．解析には機械学習を用いており，複雑な処理を行う脳機能の解明と機械学習の相性の良さを感じた．
 

発表タイトル       ：The role of multisensory signals in interval timing: An MEG study 著者                  ： T. W. KONONOWICZ, L. LECOUTRE, V. VAN WASSENHOVEセッション名           ：Human Cognition: Timing and Temporal Processing Abstruct            ：Duration perception has been shown to be influenced by the sensory modality; for instance, for the same physical durations, auditory stimuli have often been reported to be perceived as longer and more accurately than visual stimuli (Penney, Gibbon, and Meck, 2000). This auditory dominance is even present when audio-visual durations are used (Burr, Banks, Morrone, 2009; Chen and Yeh, 2009). In the context of internal clock which integrates temporal pulses to provide an estimates of elapsed time, auditory dominance is believed to result from the enhanced pacemaker rate, something that could be reflected in a build up of sustained magnetic fields (Sieroka et al., 2003) . Alternatively, duration and accuracy of temporal performance could be reflected in a pre- and pos- interval neural dynamics (Mayo and Sommer, 2013). Given that the neural signatures of multisensory duration perception have been rarely investigated, we asked participants to discriminate subsecond intervals marked by Audio, Visual, and Audio-Visual stimuli. This study was also designed to address how expectancy of a certain modality can modulate duration perception and how this expectancy changes pre-interval dynamics. To this end, participants were cued with cues that were predictive of interval modality. We will demonstrate how interval modality and cue predicting interval modality impacts subjective timing. Moreover, the role of ramping activity (Wittman, 2013) and oscillatory dynamics in duration discrimination will be discussed (Kosem, Gramfort, and Van Wassenhove, 2014).

この研究は感覚刺激の知覚時間が特定の感覚刺激の提示感覚の予測によってどのように変化するかを調べた物です．異なる感覚刺激が同じ現象であるかどうか判断する条件としてどれだけの時間間隔でそれらが生じているかという研究は有名ですが，これはその後の統合について研究しており，多感覚統合の例には様々な物があり面白いと思いました．
 
参考文献

• Neuroscience2014，http://www.sfn.org/annual-meeting/neuroscience-2014