投稿日: 投稿者: HiroyasuTomo

【速報】　The 22nd Annual Meeting of the Organization for Human Brain Mapping

2016/6/26 ~　2016/6/30の日程で、 Geneva, Switzerlandにて開催された　The 22nd Annual Meeting of the Organization for Human Brain Mapping　に研究室から２名の学生がポスター発表しました。

  • A study of performances and brain activities during memorizing tasks under the influence of sound 片山 朋香(M1)，日和 悟，廣安 知之
  • Examination of light environmental effects on brain activity using paper and computer screen 田中勇人(M2)，日和悟，廣安知之


学会参加報告書

報告者氏名 片山 朋香
発表論文タイトル
発表論文英タイトル A study of performances and brain activities during
memorizing tasks under the influence of sound
著者 片山 朋香，日和 悟，廣安 知之
主催 Organization for Human Brain Mapping
講演会名 The 22nd Annual Meeting of the Organization for Human Brain Mapping
会場 Palexpo Exhibition and Congress Centre, Geneva, Switzerland
開催日程 2016/6/26 ~　2016/6/30

 
 

  1. 講演会の詳細

2016/06/26から2016/06/30にかけて，スイスのジュネーブ（Palexpo Exhibition and Congress Centre）にて開催されましたThe 22nd Annual Meeting of the Organization for Human Brain Mappingに参加いたしました．この学会は，ヒト脳の高次機能を様々なイメージング装置によって解明するために，最新かつ革新的な研究の情報を交換することや研究成果について議論することを目的に毎年開催されています．
私は26日から30日の全日に参加いたしました．本研究室からは他に日和先生，田中勇人さんが参加しました．
 

  1. 研究発表
    • 発表概要

私は29日の午後のPoster Sessionに参加いたしました．発表の形式はポスター発表で，2時間自由に参加者の方と議論を行いました．
今回の発表は，「A study of performances and brain activities during memorizing tasks under the influence of sound」　という題目で，雑音が記憶課題時の成績と脳活動に与える影響について発表を行いました．以下に抄録を記載致します．

【Introduction】
Many people spend a lot of time to perform intellectual works such as office work or study in recent years. It is concerned that people feel stress and work efficiency is deteriorated by the effect of the surrounding environments. Since intellectual works are processed mainly through brain activities, we investigate brain activities under the influences of different sounds by using functional Near-Infrared Spectroscopy (fNIRS). The previous study reported gender differences in performances and brain functions during memorizing tasks under the influence of sounds. In this paper, we divided subjects into three groups according to the brain activities and differences of groups are discussed.
【Methods】
We researched the effect of silent environment and white noise environment on performances and cerebral blood flow (CBF) during memorizing tasks by using fNIRS. Since the operating noises and size of fNIRS are small, fNIRS is suitable for the experiment in which subjects hear sounds. 20 subjects which consisted of 10 males and 10 females participated in the experiment. As the intellectual work in this experiment, we used the task to memorize numerical strings; subjects memorized eight numbers displayed in a circle in 3 seconds, and input them in the correct order within 7 seconds. After the experiment, questionnaires were conducted to investigate subject’s psychological factors by using Visual Analogue Scale (VAS).
【Results】
The CBF changes were observed near the inferior frontal gyrus (IFG) and the middle temporal gyrus (MTG) relating to working memory or visual cognition. In these areas, the CBF changes of silent environment and those of white noise environment were compared. Subjects were classified into three groups. In the first group, there are four subjects and the CBF changes of subjects were increased under white noise environment. In the second group, there are eight subjects and the CBF changes were increased only under silent environment. In the third group, there are eight subjects and the CBF changes were increased under the both sound environment. The averages of numerical strings were calculated as score and score of silent environment and that of white noise environment were compared. It was suggested that the scores were better under the sound environment where the CBF changes were increased. From the questionnaires, the subjects who showed better scores under the white noise environment did not feel white noise unpleasant. Since white noise has the effect of masking and improving concentration power, it has a good influence for subjects who did not feel the white noise unpleasant. On the other hand, the subjects who showed better scores under the silent environment felt the white noise unpleasant. Asano et al. reported that CBF changes were decreased under unpleasant sound by using fNIRS. Since the subjects felt the white noise unpleasant, performances of them were worse and the CBF changes of them were decreased under the white noise.
【Conclusions】
In this experiment, we researched the effect of sound environment on performances and CBF changes during memorizing tasks by using fNIRS. Questionnaires were used to investigate psychological factors. Near the IFG and the MTG related to memorizing tasks were activated. The sound environment which a subject felt as pleasant was different between subjects. The performances of the subjects who did not feel the white noise unpleasant were better and CBF changes of them were increased under white noise environment. On the other hand, the performances of the subjects who felt the white noise unpleasant were worse and CBF changes of them were reduced under the white noise environment. White noise gives positive effect on people who do not feel white noise unpleasant but negative effect on people who feel white noise unpleasant. Therefore, the effect of sound environment on people is different depending on their psychological factors.

 

  • 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．先行研究において雑音環境下での記憶課題において男女差があるが，今回の研究ではどう違うのかという質問を受けました．この質問に対して，最初は男女の検討を行ったが傾向は見られなかった．そのため，ホワイトノイズの影響が良い被験者と悪い被験者がいると考え今回の研究では2群に分けたと回答しました．
 
・質問内容2
島津製作所の方から，課題の順番による影響は考慮されているのかという質問をされました． 課題の順番は被験者によって変えてカウンターバランスをとっていると回答したのですが，順番が異なる2群に対して有意差はないか確認すべきという指摘を受けました．細かい影響についてもひとつずつ影響がないか考慮する必要があると感じました．
 
・質問内容3
こちらも島津製作所の方からの質問です．2つのグループに分けていることに対して評価はしないのか，2つに分けられるとは限らないし他にグループがある可能性はないのかという質問を受けました． この質問に対して，可能性はあるが検討できていないと答えました．音の影響を受けないというグループもあるが，今回はクラスタリングの結果から大きく2グループいることから検討を行いました．他のグループについても考える必要があると思います．
 
・質問内容4
NIRSのポスターセッションが同じ方からデオシキヘモグロビンは見ていないのかという質問を受けました．これに対して検討に使っていないと回答しました．オキシヘモグロビンが最も脳活動を表していると考えられるため検討に使っていましたが，デオキシヘモグロビンを見ている研究もあるため何がわかるのか，どういうときにデオキシヘモグロビンを見るべきなのかは知る必要があると感じました．
 
・質問内容5
笹井さんという方から質問を受けました．聴覚以外の他の音に対する注意などの部位のdiscussionはどうなっているのかという質問でした．検討できていないと回答しました．聴覚や記憶に関係する部位ばかりに目がいっていたので，この指摘はとても参考になりました．今後検討したいと思います．
 

  • 感想

今回のOHBMでの発表が私にとって初めての国際学会となりました．NIRSを使って研究している方などたくさんの方が興味を持ってポスターを聞きにきてくれました．英語での説明はとても不安だったのですが，ある程度コミュニケーションがとれたのでよかったです．しかし，深く研究について話すことはできなかったため，言葉が通じ合わないと議論という段階まで到達しないということを実感しました．英語力がもっとあればさらに吸収できるものがあったと思います．また，国際学会は国内の学会と比べて規模が大きく活気にあふれていて，世界の最先端の研究について触れることができました．今後も研究を頑張っていこうと思えるいい機会になったと思います．
 
 

  1. 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       ：　Involvement of prefrontal cortex in prospective memory: An fNIRS study
著者                  ： Clarisse Aichelburg, Paola Pinti, Arcangelo Merla, Antonia Hamilton, Ilias Tachtsidis, Paul Burgess, Sam Gilbert
セッション名       ： Poster Session
Abstract            ：
Introduction: Our ability to execute delayed intentions is referred to as prospective memory. This ability plays a fundamental role in everyday behavioural organisation, and its neural correlates have been studied in a growing body of PET and fMRI studies. These studies have consistently revealed signal change within rostral prefrontal cortex [1,2]. However, to our knowledge, no previous study has investigated this ability using fNIRS. This technique presents the advantages of being portable and placing fewer constraints on participant movement [3]. Here, we investigate whether prefrontal cortex activity changes reported in previous fMRI studies examining prospective memory can also be detected using fNIRS.
Methods: Prefrontal cortex activity was monitored using a 16-channel Wearable Optical Topography (WOT, Hitachi High-technologies Corporation, Japan) fNIRS system (sampling frequency=5 Hz). 16 participants underwent four blocks of trials, during which they performed an identical ongoing task. On each trial, the participant either viewed a pair of objects or a pair of faces, and made a judgement of heaviness (i.e. selecting the heavier of the two objects, or the face belonging to the heavier person). During two Ongoing blocks, this was the only task to perform. During the Social Prospective Memory block, participants performed this task and also maintained a delayed intention to press a separate response button if they encountered a pre-specified face cue. During the Non-social Prospective Memory block, participants maintained a delayed intention to press a separate response button if they encountered a pre-specified object cue. Within each of the four blocks, ongoing trials were alternated with a low-level baseline condition. fNIRS data were first converted into haemoglobin concentration changes using the modified Beer-Lambert law. Motion artefacts were identified and corrected and a 3rd order Butterworth band-pass filter (0.01-0.5 Hz) was employed to remove physiological noise and slow drifts. For each participant and condition, signals were baseline-corrected and averaged across the four blocks; oxyhemoglobin (HbO2) and deoxyhemoglobin (HHb) peak values were computed on the averaged signals and used for statistical analyses. Grand-average was computed as well across all the participants.
Results: Separate analyses on HbO2 and HHb were performed. Channels were first identified that showed a significant response collapsed across the four conditions, in comparison with the low-level baseline (see Figure for an illustration of channel locations). Averaged across these channels, peak values were then compared between the Prospective Memory and Ongoing conditions. This showed a significant HbO2 increase during the Prospective Memory compared with Ongoing conditions. HHb analysis revealed a marginally-significant negative response associated with the Prospective Memory compared with Ongoing conditions. There were no channels showing deactivation associated with the Prospective Memory conditions. The two Prospective Memory conditions did not differ significantly from each other
Conclusions: Previous fMRI and PET studies have identified lateral prefrontal activations during prospective memory tasks along with medial prefrontal deactivations. The present fNIRS study identified activations only. These results indicate the suitability of fNIRS as a method for investigating prefrontal cortex activity during prospective memory tasks. However, this technique may be more sensitive to activations reported in previous fMRI studies rather than deactivations.

fNIRSを用いて前頭部を計測しているかたでした．簡易的なfNIRSが使われており，研究室にあるような全脳が計測できるNIRSは珍しいそうです．NIRS-SPMを使ってGLMにより活性を判断しており，やはりGLMが一般的に使われているということを実感しました．体動についてもSPMを
使って補正しているとのことでした．体動が含まれている箇所は除外している現状であるため，何かしら除去できるようにする必要があること，また，そういったアーチファクトが少なくなるような実験環境を整えることも必要だと思いました．
 
 

発表タイトル       ：FNIRS: Prefrontal activation during social vs. non-social intentions in a naturalistic setting.
著者                  ： Paul Burgess, Clarisse Aichelburg, Paola Pinti, Frida Lind, Sarah Power, Elizabeth Swingler, Arcangelo Merla, Sam Gilbert, Ilias Tachtsidis, Antonia Hamilton
セッション名       ：　Poster Session
Abstract            ： Introduction: Delayed intentions that have a pro-social aspect seem to have a special status for humans. We rate them as particularly important, and are more likely to remember to carry them out1,2,3,4. We also know that prefrontal cortex, especially area 10, plays a critical role in creating and maintaining delayed intentions5. However, it is difficult to investigate the neural substrates of social intentions in a typical neuroimaging environment since the participant is separated from other people, and there are constraints on voluntary movement which alter the ways people behave. Experimental paradigms therefore tend to be artificial in format, challenging the validity of the findings. In this experiment we used wireless fNIRS to contrast prefrontal cortex activation during the maintenance and execution of intentions relating to pro-social vs. non-social cues in a naturalistic environment. The experiment was conducted outside, on a typical London street, with participants free to move and act as they would normally.
Methods: The experiment was conducted in Queen Square, London, with no special preparation or constraint of the environment. 17 participants undertook a prospective memory (PM) task which requiring remembering to carry out intended actions after a delay period filled with a series of ongoing tasks. A typical ongoing task was finding out how many street signs there were (Figure 1A). There were 2 prospective memory conditions: social and non-social. The non-social PM task was to approach and touch any parking meters with a fist (Figure 1B). The social PM task was to approach and fist-bump a confederate who stood in predetermined locations around the square (Figure 1C). We also administered several baseline conditions to account for changes during walking, counting etc. Conditions were counterbalanced across participants and a typical timeline is shown in Figure 1D. Prefrontal cortex activity was monitored using a 16-channel Wearable Optical Topography (WOT, Hitachi High-technologies Corporation, Japan) fNIRS system (sampling frequency=5 Hz). The fiberless WOT headset was shielded from the environmental elements (rain, sunlight) with a shading cap, shown being worn by the participant in Figure 1. The engineering and procedural aspects of data collection were as described by Pinti et al6. Locations of the measurement channels relative to the brain are shown in Figure 1E.
Results: Data was preprocessed to remove artefacts and a design matrix built for each participant using the NIRS-SPM software package7. The design matrix modelled PM events and each task block separately, and was fitted for each data channel and signal type (oxy / deoxy / total Hb) separately. We considered as notable only those contrasts where independent significant changes (p<0.05) were obtained for both oxy- and de-oxy (HbO2 and HHb) hemoglobin concentration, and that these were mirror relations to each other. Relative to walking and observing the environment, when participants were also maintaining a delayed intention, there was significantly increased HbO2 with corresponding HHb change in a wide number of medial and lateral PFC regions (Fig 2 A & B). There were more specific medial rostral PFC changes (HbO2 increase; HHb decrease) relative to the ongoing task only (Fig 2C). These results are broadly in accordance with findings from fMRI and PET studies of prospective memory5,8. Most critically however, the social cues condition saw a significant increase in HbO2 (with significant decrease in HHb) in lateral prefrontal cortex regions (Fig 2D).
Conclusions: This is the first time, to our knowledge, that brain activation during a cognitive task has been measured outside the lab, with participants free to move and behave as they wish. These results suggest that the advantage for pro-social delayed intentions is reflected in differences in activation within prefrontal cortex, and that this can be detected even in naturalistic settings.

30分間外で歩き回っていくつかのタスクをしているときの脳活動を計測している研究でした．この方もNIRS-SPMを使って前頭部を計測していました．NIRSを使って動いているときの計測ができるということは知っていたのですが，座って計測を行っている場合においても体動がどうしてもデータにのってしまいます．野外で歩くときを計測するとなるアーチファクトがどうしても多くなると思いどのように処置しているのか聞いてみたのですが，このかたから説明を得られませんでした．NIRSを用いて計測をする一番のメリットというのは，日常的な状態を計測できるという点にあると改めて思いました．　NIRSにおいて光路長が計測できないという問題があるため，何かのタスクをしてどこの部位が活性しているか調べるというよりも，基準となるコントロールタスクと条件を変えたタスクとの比較をするという方法をとることが適切だと感じました．
 

発表タイトル       ：The effect of physical exercise on associative memory, a NIRS study
著者                  ： Kinga Igloi, Blanca Marin Bosch, Guido Ferretti, Aurélien Bringard1, Sophie Schwartz
セッション名       ： Poster Session
Abstract            ： Introduction: Regular physical exercise has been shown to benefit not only general health but also neurocognitive functions, especially enhancing neurogenesis in the hippocampus (Hillman, et al., 2008). A single exercise session further increases the levels of endocannabinoids, small lipophilic molecules affecting dopaminergic transmission (Tantimonaco, et al., 2014). However, little is known on the effects of a single exercise session on cognitive functions. To address this issue, here we investigated the effect of acute exercise on memory.
Methods: Sixteen healthy young participants (18 to 38 years old) performed an associative memory task twice. Each visit consisted of three parts: (1) memory encoding, (2) exercise or rest, (3) memory test (Figure 1A). We used Near InfraRed Spectroscopy (NIRS) to track changes in oxygenated hemoglobin concentration over the prefrontal cortex, during the sport or rest part and the test part. Our NIRS setup was composed of eight optodes forming seven channels. For six channels the emitters and receivers were 35mm apart, these channels were measuring effects in the brain. The remaining channel was a short separation channel with emitter and receiver 15mm apart to measure superficial layers only. We used Homer2 (Center of Biomedical Imaging, University of Harvard) to analyze NIRS data: our processing stream allowed us to identify and exclude any motion artifacts as well as to subtract the effect of superficial layers, recorded by the short separation channel from the rest of the channels.
Results: We report a significant increase in performance in the memory task after exercise, compared to after rest, which cannot be attributed to differences in vigilance (Figure 1B). This effect is paralleled by higher amplitude of the hemodynamic response function (HRF) for correct responses over area BA10 after exercise than after rest, consistent with a role of BA10 in memory processes.
Further, memory improvement between exercise and rest conditions correlated with larger deactivation over prefrontal regions of the NIRS signal during physical exercise than during rest, especially for the most difficult trials (Figure 1C). This result may suggest that prefrontal deactivation during exercise might be involved in memory consolidation, an effect that has already been shown in sleep studies (Dang-Vu, et al., 2010).
Conclusions: Overall we show that sport enhances memory consolidation, an effect which may have crucial implications in our increasingly sedentary societies. Memory improvement correlated with deactivation over prefrontal regions using NIRS during physical exercise. There might be a link to dopaminergic transmission over prefrontal cortex via endocannabinoids known to be increased by acute physical exercise.

こちらも運動中の脳活動をfNIRSで計測した研究でした．トレーニングをした後と休んでいる状態の後に記憶のテストをした脳活動を比較していました．アーチファクトを除くために3.5cmの他に1.5cmの長さのプローブ配置も行っていました．検討を行うためにきれいなデータをとるということが第一に必要なことだと思います．この方法でどこまで除去できていたかは確認できなかったのですが，しなければいけない課題だと思いました．
 

発表タイトル       ：Neural substrates of feature-based joint attention: a hyperscanning functional MRI study
著者                  ： Hiroki Tanabe, Ayumi Yoshioka, Takahiko Koike, Eri Nakagawa, Motofumi Sumiya, Shuntaro Okazaki, Norihiro Sadato
セッション名       ： Poster session
Abstract            ： Introduction: Typical joint attention is to coordinate and share spatial attention between interactive two persons regarding objects (Mundy, 2009). It is possible to extend it for object-based joint attention, and it might be a base of shared intention between two persons. It seems to be important to clarify what is core neural substrate of shared attention and what is the difference between spatial and feature-based attentional sharing during real interaction. To elucidate this, we conducted a hyperscanning functional MRI study (Koike et al., 2015) employing verbal-cued spatial and feature-based joint attention tasks.
Methods: Forty-four volunteers (22 pairs) were participated in the present study. We assigned the participants of the same gender to pairs, and they had never seen each other before the experiment. To accomplish a reciprocal live interaction of the pair, the two MRI scanners (Siemens Verio 3T) were combined with online video cameras, microphones, and headphones (Koike et al., 2016). Before the joint attention task, they chatted through the microphone and screen that showed the live video images of each participant’s face. During the task session, they looked the screen that presents visual stimuli and communicated via voice. MRI time-series data were acquired using ascending-ordered multiband EPI sequences. Each volume consisted of 36 slices, each 3 mm thick with a 0.5 mm gap, to cover the whole brain. The acquisition time was 500 ms and the delay-in-repetition time was 2,000 ms. This 2 sec is silent (no scanner noise) and participants talked each other within this period. We acquired 175 volumes per run. Image preprocessing and statistical analysis was performed using SPM12. In the feature-based joint attention task, the object(s) were displayed at the center of the screen (Fig.1a). It has 4 DIMENSIONS such as NUMBER (1, 2, 3, or 4), SHAPE (star, heart, circle, or square), COLOR (red, blue, yellow, or green), and PATTERN (stripe, border, check, or dot). The roles of initiator and responder were noticed by the color (cyan or magenta) around the edge of the screen. The initiator freely chose one of four dimensions and informed it by utterance. For example, he/she said ‘SHAPE’ during the silent period. The responder listened to the initiator’s word, and attended the same dimension of the object. Next, he/she replied the characteristic of this dimension. For example, he/she said ‘star’ during the next silent period. The initiator judged whether it is correct or not and feedback verbally in the same silent period.
In the spatial joint attention task, four target objects were displayed at the above, below, left, and right from the center of the screen (Fig.1b). The initiator freely chose one of four objects and informed the place of it (i.e. above, below, left, or right) by utterance. The responder listened to the initiator’s word, and looked at the same object. Then, he/she replied the shape of the object. The initiator judged whether it is correct or not and feedback verbally. We also set 2 control conditions those were done solitary.
Results: The results showed that bilateral inferior frontal region, anterior insula, superior temporal gyrus, pre-supplementary motor area, anterior cingulate cortex, medial prefrontal cortex, precuneus, and thalamus were commonly activated in both spatial and feature-based joint attention tasks. Feature-based joint attention specific regions were bilateral occipital pole, thalamus, anterior cingulate cortex, and left inferior frontal gyrus. Spatial joint attention specific regions were bilateral frontal and supplementary eye field, lateral occipital cortex, lingual gyrus, and cuneus.
Conclusions: We identified core neural substrates for shared intention. To consider the results of our previous studies (Tanabe et al., 2012; Koike et al., 2016) together, the inferior frontal region and the anterior insula might play a significant role in communication with sharing attention during live social interaction.

MRIを2台使って2人の同時計測をしている研究でした．その研究ができる環境が本当に貴重だと思いました．私自身，学生のうちに高価な医療機器を使える環境で研究ができているので大変恵まれていることだと改めて感じました．2人を同時に計測するという研究は今までに論文や先輩の研究で知ってはいたのですが，今回お話を聞いてコミュニケーションの研究の面白さに気づくことができました．同時計測するときには同じ課題を一緒にするという認識をしていたのですが，この研究では質問する人と回答する人という異なる役割に対して計測をしていました．2人を同時に計測するからこそできる実験なのがとてもいいと思います．計測している組み合わせによって結果は異なるというのはさらに難しいことだと思いますが，ヒトを対象として実験することに対してのおもしろさが詰まっている部分だとも思います．fMRIの空間分解能の高さがあるため，同じ活性部位内であっても異なる役割で細かくみると位置が少し異なるという点が非常に興味深かったです．NIRSで同時計測することによって，細かくみるということは難しいですがより日常に近い状態で計測することが可能であるため，機会があれば実験してみたい内容だと思いました．
 

発表タイトル       ：Real-time motion artifact removal algorithm for fNIRS to offset head motion using motion sensors
著者                  ： Jae-Myoung Kim, Jong-Kwan Choi, Min-Gyu Choi, Gunpil Hwang, Minsu Ji, Hyeon-Min Bae
セッション名       ： Poster session
Abstract            ： Introduction: Functional near-infrared spectroscopy (fNIRS) is one of the non-invasive functional brain imaging techniques offering portability and high resolution [1]. However, fNIRS is prone to motion artifacts (MAs) caused by external forces influencing the brain hemodynamics. Examples include hemodynamic baseline drift (HBD) due to head motions. A recent study shows that the extent of drift is non-linear and it is indeterminable with motion sensor measurements [2]. Therefore, several post processing methods to remove MAs after fNIRS data acquisition have been reported [3]. In this paper, the relationship between the head angle and the extent of HBD is investigated and accordingly, a real-time MA removal algorithm for an fNIRS system measuring prefrontal cortex is provided.
Methods: Proposed model :The blood circulation in a human body is influenced by gravity. Cerebral venous pressure (CVP) is at a minimum when a person is sitting up straight as given in Fig.1-(1) [4]. When a person tilts one’s head down, the variations in the CVP can be modeled as described in Fig.1-(2, 3). Moreover, the amount of venous side cerebral blood volume (CVV) shifting is linearly proportional to the compliance of cerebral vein under CVP changes [5, 6] (see Fig.1-(4)). Such CVV shift eventually causes changes in the optical signal intensity in an fNIRS system [7].
Data acquisition: Changes in the optical intensity (OI) were measured by NIRSIT (OBELAB, Rep. of Korea), a continuous wave fNIRS system that utilizes lasers with wavelengths of 780 nm and 850 nm and a 6-axis motion sensor that monitors the head movement of a test subject simultaneously. As shown in Fig. 2(a), a total of 48 channels with a source-detector separation of 3cm were measured in the prefrontal cortex. The center of the lowermost probes was aligned at the FPz in the 10–20 EEG system to remove positional uncertainty among subjects.
Experiment procedure: For the experiment, a total of four healthy adult (age-range 22 to 28; mean, 24.4; n=1 women) subjects were recruited. For each subject, Type A protocol was performed initially, as shown in Fig. 2(b) and then, Type B protocol illustrated in Fig. 2(b) was repeated 10 times while maintaining identical probe arrangement. In the Type B protocol, subjects were asked to tilt their head forward at a random degree in three steps and maintain their head position for 20 seconds for each step.
Data analysis: The OI variation in each channel was sampled at 8.138 Hz, and the angle changes ‘θ’ were traced in real-time by using a gyroscope and an accelerometer, wherein complementary filtering is employed [8]. The OI variation was filtered by using an 8-tap moving average filter and then converted to variations in the optical density (OD) [7]. Individual fitting parameter ‘G’ in Fig. 3(a) was obtained through Type A protocol and then applied to the proposed MA removal algorithm illustrated in Fig. 3(b) for evaluation during Type B protocol. Non-linear regression analysis was performed on the observed data with respect to equation in Fig. 3(a)-(1), where the channels with insufficient SNR, due to hair and/or detachment of probes during the protocols, were ignored. Statistical analysis was performed using GraphPad Prism version 5.
Results: Fig. 4 shows a strong correlation between the analytic estimations and the observed data. The regression was significant at each channel, resulting in R²=0.33~0.91 for all subjects as shown in Fig. 4. These results validate that the OD change under head tilting can be modeled by the formula proposed in Fig. 3(a)-(1).
Conclusions: The variations in the OD under head tilting procedure can be estimated analytically and be removed in real-time effectively by using the proposed MA removal algorithm. Individual fitting parameters can be retrieved at startup via a single gain calibration procedure. The proposed user-characterized MA removal algorithm in Fig. 3(b) presents the possibility for use in modern cognitive neuroimaging research.

頭の傾きによるアーチファクトを除去するアルゴリズムの研究している方でした．細かい方法については理解することができませんでした．頭の傾きによって圧力がかかり，ベースラインドリフトがアーチファクトトとして生じてしまいます．頭の傾きが血流移動に影響していることは聞いたことがあったのですが，実験時に気に留めたことがあまりなかったので注意すべき点だと気づきことができました．モデルとデータがあっているかも検証しており，そういったモデルをたてるという方法も試してみるべきだと思いました．
 
参考文献

  • OHBM 2016 Annual Meeting,

http://www.humanbrainmapping.org/i4a/pages/index.cfm?pageID=3662
 
学会参加報告書

報告者氏名 田中勇人
発表論文タイトル 照明環境が紙媒体とディスプレイ使用時の脳活動に与える
影響の検討
発表論文英タイトル Examination of light environmental effects
on brain activity using paper and computer screen
著者 田中勇人，日和悟，廣安知之
主催 Organization for Human Brain Mapping
講演会名 The 22nd Annual Meeting of the Organization for Human Brain Mapping
会場 Palexpo Exhibition and Congress Centre
開催日程 2016/6/26-2016/6/30

 
 

  1. 講演会の詳細

2016/6/26から2016/6/30にかけて、Palexpo Exhibition and Congress Centre (Geneva)にて開催されましたThe 22nd Annual Meeting of the Organization for Human Brain Mappingに参加いたしました．この学会にはヒトの脳機能イメージングの分野で活躍される研究者が数多く参加され，最新かつ画期的な研究内容において意見交換を行うことを目的に開催されました．
本研究室からは日和先生，片山さん，私が開催期間全日に参加し，片山さんと私がポスター発表を行いました．
 

  1. 研究発表
    • 発表概要

私は，29日の12:45-14:45のポスターセッションにて発表を行いました． ポスターセッションの120分間は，ポスター内容に興味を示してくださる方々から質問をいただき，その内容に関してディスカッションする時間でした．
以下に抄録と質疑応答の内容を記載いたします．

【Introduction】 In this study, the best light environment to improve intelligent productivity of office workers is examined. It is reported that light environmental effects on their task performance [1][2]. Working efficiency is processed in the brain, and it can be evaluated quantitatively by testing the neurobehavioral functions, including perception, learning, memorizing and thinking [3]. We have considered the effect of light environment on brain activities during visual task and this visual task was provided using paper display media [4]. However, office employees often perform this task using paper and computer screen. We investigated the effect of light environment on brain activity during visual task using paper and computer screen.
【Methods】 In this experiment, three types of light environment were used; Red, White and Blue. The illuminance condition was maintained at 700 lx and combined with three different color temperature conditions (Red: 3000 K, White: 4500 K, Blue: 7000 K). 15 subjects performed a visual search task using paper and computer screen under Red, White and Blue lights. They were instructed to determine whether there is “T” among a lot of “L” as soon as possible. They repeated this task for 90 seconds. In each type of light, the CBF (cerebral blood flow) changes in the frontal lobe (22ch) and temporal lobe (48ch) were measured by fNIRS (functional near-infrared spectroscopy). Task performance was evaluated as percentage of correct answers without errors. When subjects made mistakes or could not react in a certain time interval, this reaction was regarded as an error. To examine the level of brain activity, the average value of oxy-Hb during task interval was calculated.
【Results】 There was the tendency that some subjects got much better task performance in Blue than in White when using paper and they got much better performance in Red when using a computer screen. There was the significant difference of task performance using paper between Blue and White at the 5% level, and there was the significant difference of performance using computer screen between Red and White. The region whose average value of oxy-Hb changes is higher than the total region is regarded as the active region. In Blue, CBF change increased using paper at the fusiform area. In Red, fusiform area was active using the computer screen. Fusiform area works for visual perception function. Therefore, it was found that the activation of this area led to the improvement of visual performance. Compared to the result of task performance in the computer screen, task performance with paper was better in Red or White. Computer screen emits the transmitted light. Therefore, it was considered that the light led to the deterioration of visual performance in Red and White.
【Conclusions】 The purpose of this study is to investigate the effect of light environment on brain activity during visual task. In this paper, the brain activity under Red, White and Blue environment using paper and computer screen is measured, and the differences of brain activity are discussed. In this experiment, the task performance when using paper was best in Blue significantly, and the one using the computer screen was best in Red. By examination of brain activity by fNIRS, the activation of the fusiform area which has visual perception function was observed in circumstances where visual performance was the best. The transmitted light from the computer screen may effect on visual performance differently, if light environment is changed.
【References】
[1] K. Smolders, Y. Kort and P. Cluitmans (2012), ‘A higher illuminance induces alertness even during office hours：Findings on subjective measures, task performance and heart rate measures,’ Physiology and Behavior, vol. 107, no. 1, pp. 7-16．
[2] P. c. Changa, S. y. Choua and K. k. Shiehb (2013), ‘Reading performance and visual fatigue when using electronic paper displays in long-duration reading tasks under various lighting conditions,’ vol. 34, issue 3, pp. 208-214.
[3] L. Lan, Z. Lian, L. Pan and Q. Ye (2009), ‘Neurobehavioral approach for evaluation of office workers’ productivity: The effects of room temperature,’ Building and Environment, vol. 44, no. 8, pp. 1578-1588.
[4] H. Tanaka, T. Hiroyasu (2015), ‘Examination of the effect of light environment on brain activity during visual search task by fNIRS,’ Neuroscience.

 

  • 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
今回の学会での質疑応答では、fNIRSデータ処理よりも実験概要や結果に関して説明を求められることが多かったです．質問に来てくださった方は，ポスターセッションでfNIRSを用いた研究に関して発表を行った研究者でした．
University College LondonのPaul BurgessさんやClarisse Aichelburgは，「タスクとレストのOxy-Hb平均値の差を標準化したCohen’Dで活性を評価しているのは興味深いけど，Deoxy-Hbデータを用いて評価は行っているの」という質問をいただきました．この質問に関しては，「目視でDeoxy-Hbの波形はみたものの，今回はOxy-Hbデータのみで活性チャンネルを検討しました．」とコメントしました．
また、一通りの実験結果を説明した後に，「ここまで照明環境によって活性部位が異なるということは，おそらく機能的コネクティビティにも周囲の環境は影響を与えているのかもしれない」というコメントを，複数の方からいただきました．このコメントをいただいた方のお名前を控えるのを忘れてしまいました．
 
 

  • 感想

今回は，二回目の国際学会参加で，ポスター発表においては国内外合わせて四回目であったことから，発表自体にそこまで緊張することはありませんでした．しかし，自身の英語で研究内容が伝わるのかという不安はありました．英語できちんと発表できるように練習をして発表に臨んだので，結果的に自分の言いたいことは言えたのではないかと思います．私のポスターに興味を示していただいた方のおっしゃっている内容も，大まかには理解することができました．ディスカッションをしていて感じたことは，日本の解析は海外と比べると遅れを取っているということです．海外の学生は，fNIRSデータを用いて機能的コネクティビティの変化に関して検討を行ったり，GLMやSPMを用いて活性指標を作り上げたりしていました．私もあと約半年の研究生活において，このような解析を自身の取得データに適用して，納得のいく修士論文を書き上げていきたいなと，今回の国際学会に参加して強く感じました．
 
 

  1. 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 ： Altered Resting State Functional Connectivity in Mild Cognitive Impairment revealed by NIRS
著者 ： Andrei Medvedev, Raymond Turner
セッション名 ： Poster Session (Monday, June 27 2016: 12:45 PM – 02:45 PM)
 
Abstract
Introduction: Treatment of Alzheimer’s disease (AD) is hampered by our limited ability to detect the disease early. Degenerative alterations in neuronal branches disrupt normal communication within and between functional neuronal networks widely distributed over the cortex. Functional connectivity (FC) may serve as an indicator of pathological processes in the brain because FC patterns experience significant changes in several neurological and psychiatric disorders (Seeley et al., 2009; Cabral et al., 2014). Functional MRI has been successfully used to study functional brain networks. However, there are limitations in the use of fMRI related to cost, sensitivity to motion, claustrophobia and presence of metallic objects in the patient’s body. We used near-infrared spectroscopy (NIRS) to study FC in individuals with Mild Cognitive Impairment (MCI). NIRS is a promising imaging technology in being cost-effective, less restrictive, relatively insensitive to motion artifact and suitable for use in virtually all subjects.
Methods: We recorded high density NIRS signals (16 x 32 dual wavelength source-detector pairs) in five MCI subjects and four age-matched healthy participants during the resting state. Data preprocessing included the standard steps for artifact removal using Independent Component Analysis (Medvedev et al., 2008) and bandpass filtering 0.01-0.1 Hz. Methods for calculation of functional connectivity were as follows (Medvedev et al., 2011): channel-wise correlation coefficients were computed and their statistical evaluation was performed by Fisher’s transform and two-sample t-tests between groups for each channel-pair.
Results: Group-averaged raw connectivity matrices show all 60 x 60 pairwise channel correlations grouped according to their anatomical labeling. These matrices reveal higher connectivity within prefrontal cortex (PFC) in controls versus MCI subjects. On the other hand, MCI participants showed increased connectivity in a few areas preferentially in the right hemisphere (marked by ovals in the MCI connectivity matrix). Overall, the MCI connectivity matrix showed a weaker but slightly more spread out pattern that may be characterized as more diffuse and with less modularity.
Conclusions: Collectively, these data suggest that NIRS is sensitive to detect the aberrant patterns of FC at the early stage of MCI due to Alzheimer’s disease. These results are promising for the development of novel biomarkers for the early diagnosis of MCI and AD as well as for the quantitative assessment of treatments and rehabilitation strategies for patients with neurodegenerative disorders.

この研究では、軽度認知障害 (MCI)と健常者においてfNIRS実験を行い，前頭前野における機能的コネクティビティ (FC)の差異を検討されていました．今後，私も照明環境がFCに与える影響を検討しようと考えているため，解析方法などを参考にしていこうかと思いました．また，fNIRSデータの前処理で，独立成分分析とバンドパスフィルタリング処理を適用されている点も非常に興味深かったです．この発表者は，チャンネル毎に相関係数を算出した後にフィッシャー変換を行ってFCを算出しているようでしたので，この方法をまず自身のデータに適用できるように勉強して，個人ごとの環境間のFCの違いを検討していこうと思いました．
 

発表タイトル ：Involvement of prefrontal cortex in prospective memory: An fNIRS study
著者 ： Clarisse Aichelburg, Paola Pinti, Arcangelo Merla, Antonia Hamilton, Ilias Tachtsidis, Paul Burgess, Sam Gilbert
セッション名 ： Poster Session (Monday, June 27 2016: 12:45 PM – 02:45 PM)
 
Abstract
Introduction: Our ability to execute delayed intentions is referred to as prospective memory. This ability plays a fundamental role in everyday behavioural organization, and its neural correlates have been studied in a growing body of PET and fMRI studies. These studies have consistently revealed signal change within rostral prefrontal cortex. However, to our knowledge, no previous study has investigated this ability using fNIRS. This technique presents the advantages of being portable and placing fewer constraints on participant movement. Here, we investigate whether prefrontal cortex activity changes reported in previous fMRI studies examining prospective memory can also be detected using fNIRS.
Methods: Prefrontal cortex activity was monitored using a 16-channel Wearable Optical Topography (WOT, Hitachi High-technologies Corporation, Japan) fNIRS system (sampling frequency=5 Hz). 16 participants underwent four blocks of trials, during which they performed an identical ongoing task. On each trial, the participant either viewed a pair of objects or a pair of faces, and made a judgement of heaviness (i.e. selecting the heavier of the two objects, or the face belonging to the heavier person). During two Ongoing blocks, this was the only task to perform. During the Social Prospective Memory block, participants performed this task and also maintained a delayed intention to press a separate response button if they encountered a pre-specified face cue. During the Non-social Prospective Memory block, participants maintained a delayed intention to press a separate response button if they encountered a pre-specified object cue. Within each of the four blocks, ongoing trials were alternated with a low-level baseline condition. fNIRS data were first converted into haemoglobin concentration changes using the modified Beer-Lambert law. Motion artefacts were identified and corrected and a 3rd order Butterworth band-pass filter (0.01-0.5 Hz) was employed to remove physiological noise and slow drifts. For each participant and condition, signals were baseline-corrected and averaged across the four blocks; oxyhemoglobin (HbO2) and deoxyhemoglobin (HHb) peak values were computed on the averaged signals and used for statistical analyses. Grand-average was computed as well across all the participants.
Results: Separate analyses on HbO2 and HHb were performed. Channels were first identified that showed a significant response collapsed across the four conditions, in comparison with the low-level baseline. Averaged across these channels, peak values were then compared between the Prospective Memory and Ongoing conditions. This showed a significant HbO2 increase during the Prospective Memory compared with Ongoing conditions. HHb analysis revealed a marginally-significant negative response associated with the Prospective Memory compared with Ongoing conditions. There were no channels showing deactivation associated with the Prospective Memory conditions. The two Prospective Memory conditions did not differ significantly from each other.
Conclusions: Previous fMRI and PET studies have identified lateral prefrontal activations during prospective memory tasks along with medial prefrontal deactivations. The present fNIRS study identified activations only. These results indicate the suitability of fNIRS as a method for investigating prefrontal cortex activity during prospective memory tasks. However, this technique may be more sensitive to activations reported in previous fMRI studies rather than deactivations.

この発表では、fNIRSデータ処理に関して質問したくて，ポスター発表を聴講しに行った．前処理としてバンドパスフィルタ処理を適用して，その後にベースライン処理を行っていることは理解できたが，その後どのような値を用いて活性チャンネルを視覚化しているのかを理解することができませんでした．結果はfMRIを用いた先行研究が示すものになったとおっしゃっていたけど，前頭前野における機能と行動データとの関連性に関する考察を質問しても，あまり納得のいく返事は返ってこなかったので，先行研究を読んでみようかと思いました．認知と記憶を測る課題を用いて社会性の違いを検討されていたので，この課題においても興味深いなと感じました．
 

発表タイトル ：Simultaneous EEG-fMRI and EEG-fNIRS reveal differences between movement execution and imagination
著者 ： Catharina Zich, Maarten De Vos, Cornelia Kranczioch, Ling-Chia Chen, Stefan Debener
セッション名 ： Poster Session (Monday, June 27 2016: 12:45 PM – 02:45 PM)
 
Abstract
Introduction: According to the neural simulation of action theory, movement execution (ME) and imagination (MI) of the same movement are different stages on a continuum and are therefore based on similar neuronal networks (Jeannerod, 2001). MI may facilitate experience-driven cortical reorganization and functional restoration, even in the absence of residual voluntary limb movement. Moreover, it has been shown that neurofeedback enhances task-specific activity and MI learning. Despite the clinical importance, it remains not known which neuroimaging technique adequately captures cortical lateralization induced by ME and MI, the latter being the basis for most MI neurofeedback applications. The present investigation therefore compared ME and MI induced lateralization using concurrent EEG-fMRI and concurrent EEG-fNIRS recordings.
Methods: Two datasets were analyzed. In both experiments MI naïve young healthy adults performed one block of ME and three blocks of MI. The first experiment (N = 22, mean age: 23.9 years) comprised two sessions, one simultaneous EEG-fMRI and one EEG session. As ME execution corrupted the EEG signal obtained inside the fMRI scanner beyond recovery, only the ME EEG data obtained outside the scanner were considered. This was justified because, as reported previously, the MI induced ERD of offline corrected EEG data and EEG data obtained outside the MRI scanner were significantly correlated (Zich et al., 2015). The second experiment (N = 19, mean age: 24.4 years) comprised one session of simultaneous EEG-fNIRS. For each experiment and imaging modality separate repeated measures 2 x 2 ANOVAs with the factors condition (execution, MI) and lateralization (contralateral, ipsilateral) were performed, and in case of significant interaction followed up with t-test.
Results: fMRI BOLD activity was more lateralized during ME than during MI. Descriptively, this pattern is confirmed by both fNIRS measures, oxygenated (HbO) and deoxygenated (HbR) hemoglobin concentration changes, but only for HbO was the effect of lateralization significant. In contrast, electrophysiological activity, specifically ERD, was found to be more lateralized during MI than during ME in both datasets.
Conclusions: Both studies revealed a clear consistent dissociation between hemodynamic and electrophysiological signatures regarding ME and MI induced cortical lateralization. Future work investigating this dissociation on EEG source level seems important to better understand the patterns of cortical activation induced by MI supported by neurofeedback. This would help to advance MI neurofeedback applications towards use in motor recovery after stroke.

この発表では、運動早期(MI)をフィードバックするアプリケーション開発の基礎研究として， EEG-fMRIそしてEEG-fNIRSの同時計測を行い，ME(運動の実行)とMIにより誘発される定位を比較していました．EEGやMRIの処理方法に関しては十分に理解することができませんでしたが，二種類の同時計測で取得したデータを用いている実験設計が個人的にすごく面白く感じました．
 

発表タイトル ： The effect of physical exercise on associative memory, a NIRS study
著者 ： Kinga Igloi, Blanca Marin Bosch, Guido Ferretti, Aurélien Bringard, Sophie Schwartz
セッション名 ： Poster Session (Tuesday, June 28 2016: 12:45 PM – 02:45 PM)
 
Abstract
Introduction: Regular physical exercise has been shown to benefit not only general health but also neurocognitive functions, especially enhancing neurogenesis in the hippocampus (Hillman, 2008). A single exercise session further increases the levels of endocannabinoids, small lipophilic molecules affecting dopaminergic transmission (Tantimonaco, et al., 2014). However, little is known on the effects of a single exercise session on cognitive functions. To address this issue, here we investigated the effect of acute exercise on memory.
Methods: Sixteen healthy young participants (18 to 38 years old) performed an associative memory task twice. Each visit consisted of three parts: (1) memory encoding, (2) exercise or rest, (3) memory test. We used Near InfraRed Spectroscopy (NIRS) to track changes in oxygenated hemoglobin concentration over the prefrontal cortex, during the sport or rest part and the test part. Our NIRS setup was composed of eight optodes forming seven channels. For six channels the emitters and receivers were 35mm apart, these channels were measuring effects in the brain. The remaining channel was a short separation channel with emitter and receiver 15mm apart to measure superficial layers only. We used Homer2 (Center of Biomedical Imaging, University of Harvard) to analyze NIRS data: our processing stream allowed us to identify and exclude any motion artifacts as well as to subtract the effect of superficial layers, recorded by the short separation channel from the rest of the channels.
Results: We report a significant increase in performance in the memory task after exercise, compared to after rest, which cannot be attributed to differences in vigilance. This effect is paralleled by higher amplitude of the hemodynamic response function (HRF) for correct responses over area BA10 after exercise than after rest, consistent with a role of BA10 in memory processes.
Further, memory improvement between exercise and rest conditions correlated with larger deactivation over prefrontal regions of the NIRS signal during physical exercise than during rest, especially for the most difficult trials. This result may suggest that prefrontal deactivation during exercise might be involved in memory consolidation, an effect that has already been shown in sleep studies (Dang-Vu, 2010).
Conclusions: Overall we show that sport enhances memory consolidation, an effect which may have crucial implications in our increasingly sedentary societies. Memory improvement correlated with deactivation over prefrontal regions using NIRS during physical exercise. There might be a link to dopaminergic transmission over prefrontal cortex via endocannabinoids known to be increased by acute physical exercise.

この発表では，運動後に記憶課題を課した場合とリラックスした後に記憶課題を課した場合の脳活動をfNIRS実験により比較していた点が興味深かったです．課題前に行った運動が， BA10(前頭極)の血流上昇あるいはドーパミン作用に影響を与えてタスクパフォーマンスが向上したという考察も面白かった．しかしそれ以上に，記憶力がレスト-タスク間で改善した被験者は，運動時に前頭前野の活性が見られなかったという結果が興味深かった．記憶に関わる脳部位がレスト区間で機能していない方がタスクパフォーマンスの向上につながる可能性も考慮して，前頭極の機能に関しても文献調査をしていこうと思いました．
 

発表タイトル ： Early development of visual working memory: an fNIRS study
著者 ： Lourdes Delgado Reyes, Sobanawartiny Wijeakumar, Vincent Magnotta, John Spencer
セッション名 ： Poster Session (Tuesday, June 28 2016: 12:45 PM – 02:45 PM)
 
Abstract
Introduction: Visual working memory (VWM) is a core cognitive system with a highly limited capacity. These limitations have a profound impact on the development of a broad range of cognitive abilities (Conway et al., 2003). The present study investigated the neural correlates of the development of VWM in early development using Functional Near-Infrared Spectroscopy (fNIRS) and the preferential looking VWM task.
Methods: Two-year-olds (n=21) completed a modified version of the preferential looking task while we recorded neural activity using a 36-channel fNIRS system with sources (s) and detectors (d) placed over frontal, temporal and parietal cortex bilaterally. Infants were presented with two side-by-side flickering displays composed of an array of colored squares. In one display, one of the items changed after a brief delay (250ms). The number of items in the array varied across trials (2, 4, 6 items). NIRS data were processed using HomER2 (Huppert et al., 2009). Channels with very low optical density were discarded. Targeted Principal Component analyses (Yücel et al., 2014) was used to correct motion. The hemodynamic response was modeled with regressors that captured stimulus timing and duration for the three conditions of interest as well as nuisance regressors. This resulted in a β value for each channel for each condition for both HbO and HbR. Each participant’s digitized optode location was registered to an average age-matched structural scan using AtlasViewer software. The probability distributions of photon migration for each s-d pair was computed using Monte Carlo simulations (Fang et al., 2009; Selb et al., 2014). These distributions were combined with GLM-based analysis of the fNIRS data to reconstruct a functional image for each participant, moving from channel space to voxel space (Wijeakumar et al., 2016). This yielded an estimate of the localized changes in Hb concentration within each voxel of the brain. Reconstructed functional images were aligned to MNI space using an affine transformation. An intersection mask was then computed across participants. These data were analyzed at the group level in AFNI using ANOVA with family wise correction (α=.05) with subject and SS (2,4,6) as factors.
Results: The time participants looked at each display was recorded online across each trial. Infants looked longer as the SS increased (F(2,40)=3.536, p<.05). To evaluate differential looking between the arrays, a change preference (CP) score was computed (change/[change + no-change]). Figure 2b shows the participants’ CP as the set size (SS) varied from two to six items. Participants preferred the changing stream at SS2, t(20)=3.088, p < 0.01, but not at the higher set sizes. This likely reflects the shorter trial durations used here relative to previous work (Ross-Sheehy et al., 2003). A central question of this study was whether infants would show task-related activity in the previously identified VWM fronto-parietal network (Todd et al., 2004, 2005). This was the case: channels across frontal and parietal regions showed task-related activation across all three SS. Moreover, results revealed an increase in activation as the working memory load increased from 2 to 4 items within the VWM network that included bilateral frontal (MFG, SFG) and left parietal (AG, IPL) regions, and a decrease when the load was further increased to 6 items (see fig. 3 and 4). Correlations between brain activity and behavior revealed that left parietal activation at SS4 was associated with the total looking time at SS4 (r(16)=0.77, p<.001).
Conclusions: Participants engaged frontal and parietal regions previously reported as part of the fronto-parietal network underlying VWM. Looking behavior was associated with activity in left IPL at SS4. Interestingly, even though participants did not show a CP at SS4, robust neural activity was observed on this condition. These results provide a first-look at the neural processes that underlie the early development of VWM.

この発表では，fNIRSデータを用いて，前頭部と頭頂部のネットワークについて検討している点が興味深かったです．活性判定はGLMを用いて解析しているようでした．GLMに関する解析手法の詳細は質問して色々おっしゃっていただけたのですが，今ひとつ理解することができませんでした．私も今後，機能的ネットワークに関する解析をしていこうと考えているので，この研究内容を参考にしてみようかと思いました．また，前頭部と頭頂部の機能的ネットワークに関しても，調べていこうと思います．
 
 
参考URL

  • http://www.humanbrainmapping.org/i4a/pages/index.cfm?pageID=3662