Kawashima I, Kumano H. Front Hum Neurosci. 2017 Jul 12;11:365. doi: 10.3389/fnhum.2017.00365. eCollection 2017.

Mind-wandering (MW), task-unrelated thought, has been examined by researchers in an increasing number of articles using models to predict whether subjects are in MW, using numerous physiological variables. However, these models are not applicable in general situations. Moreover, they output only binary classification. The current study suggests that the combination of electroencephalogram (EEG) variables and non-linear regression modeling can be a good indicator of MW intensity. We recorded EEGs of 50 subjects during the performance of a Sustained Attention to Response Task, including a thought sampling probe that inquired the focus of attention. We calculated the power and coherence value and prepared 35 patterns of variable combinations and applied Support Vector machine Regression (SVR) to them. Finally, we chose four SVR models: two of them non-linear models and the others linear models; two of the four models are composed of a limited number of electrodes to satisfy model usefulness. Examination using the held-out data indicated that all models had robust predictive precision and provided significantly better estimations than a linear regression model using single electrode EEG variables. Furthermore, in limited electrode condition, non-linear SVR model showed significantly better precision than linear SVR model. The method proposed in this study helps investigations into MW in various little-examined situations. Further, by measuring MW with a high temporal resolution EEG, unclear aspects of MW, such as time series variation, are expected to be revealed. Furthermore, our suggestion that a few electrodes can also predict MW contributes to the development of neuro-feedback studies.

Tomita N, Imai S, Kanayama Y, Kawashima I, Kumano H. Percept Mot Skills. 2017 Jun;124(3):703-720. doi: 10.1177/0031512517700054. Epub 2017 Mar 27.

While dichotic listening (DL) was originally intended to measure bottom-up selective attention, it has also become a tool for measuring top-down selective attention. This study investigated the brain regions related to top-down selective and divided attention DL tasks and a 2-back task using alphanumeric and Japanese numeric sounds. Thirty-six healthy participants underwent near-infrared spectroscopy scanning while performing a top-down selective attentional DL task, a top-down divided attentional DL task, and a 2-back task. Pearson's correlations were calculated to show relationships between oxy-Hb concentration in each brain region and the score of each cognitive task. Different brain regions were activated during the DL and 2-back tasks. Brain regions activated in the top-down selective attention DL task were the left inferior prefrontal gyrus and left pars opercularis. The left temporopolar area was activated in the top-down divided attention DL task, and the left frontopolar area and left dorsolateral prefrontal cortex were activated in the 2-back task. As further evidence for the finding that each task measured different cognitive and brain area functions, neither the percentages of correct answers for the three tasks nor the response times for the selective attentional task and the divided attentional task were correlated to one another. Thus, the DL and 2-back tasks used in this study can assess multiple areas of cognitive, brain-related dysfunction to explore their relationship to different psychiatric and neurodevelopmental disorders.

☆書誌：　人間科学研究，2012

書誌： J Phys Fitness Sports Med ,2012

Masaki H. & Sommer W. (2012). J Phys Fitness Sports Med, 1(3): 369-380. Abstract   In this review, focus is given to the cognitive brain functions associated with motor learning and the control of learned motor behavior, as revealed by non-invasive studies in humans. After providing a definition of motor control and learning, the tasks adopted in previous studies are first introduced, and some important findings about motor behavior and pertinent theoretical models are described. Relying mainly on findings from the event-related potential (ERP) technique, but also from neuroimaging, this review focuses on motor learning and motor control in skilled action, with an emphasis on movement preparation and performance monitoring. Keywords : event-related potentials, motor control, motor learning, movement preparation,  performance monitoring

書誌： 第27回日本人工知能学会全国大会 ,2013

田和辻可昌, 小島一晃, 松居辰則：視床と扁桃体の機能に着目した不気味の谷発生メカニズムのモデル化の試み, 第27回日本人工知能学会全国大会 2F1-2 (2013)

Abstract:In the research field of Human-Computer Interaction, the uncanny valley is a critical issue in designing an appropriate appearance of a humanlike agent. Although human familiarity toward an agent increases as the agent gets more similar to human, it drastically decreases to the bottom of a valley when the agent is considerably similar to real human. An agent highly similar to human is first perceived as a human and subsequently as a non-human. We propose a model of how possible a conflict of these information occurs, based on the dual-pathway of emotion and the function of the amygdala. According to the model, the conflict can the most likely occur where an agent looks highly but imperfectly human.

書誌：第29回リハ工学カンファレンス講演予稿集,2014

畠山卓朗・本間一秀・渡辺崇史・福田暁子（2014）肢体不自由を伴う盲ろう者に対するコミュニケーション支援機器の試作，第29回リハ工学カンファレンス講演予稿集

書誌： Clinics and Practice , 2013

S.Tanabe, E.Saitoh, K.Ohtsuka, T.Teranishi, Y.Tomita, and Y.Muraoka, 2013. Simple method  to reduce the effect of patient positioning variation on three-dimensional motion analysis  during treadmill gait. Vol.3, No.2, 84-86

Abstract

Recently, three-dimensional (3D) closed curve trajectories of markers placed at strategic body locations, called  cyclograms or Lissajous-like graphs, are used for treadmill gait analysis. A simple method is presented to reduce the effect of patient positioning variation. After breaking down movement into three components (anterior-posterior, medial-lateral and superior-inferior), the time-series data and time-inverted data are serially concatenated. A fast Fourier transform (FFT) is done, and a high-pass filter (except 0 Hz) is applied to the anterior-posterior and medial-lateral components. Next an inverse FFT is executed, and the posterior half of the outcome, corresponding to time-inverted data, is deleted. The 3D closed curve is then reconstructed. Results showed that the proposed method was able to reduce the effect of patient positioning variation. Since the adjusted curve is simply a symbolized gait pattern, the method might be useful as an adjunct tool in observational gait analysis.

書誌： 日本知能情 報ファジィ学会論文誌(「知能と情報」） ,2014

宮澤 幸希, 白勢 彩子, 馬塚 れい子, 菊池 英明, "知識獲得モデルとして の自己組織化マップ－連続音声からの教師なし音素体系の学習－", 日本知能情 報ファジィ学会論文誌(「知能と情報」）, vol.26, no.1，pp.510-520，2014．

要約：

人の知識獲得のプロセスを計算論的に検証する目的で，大脳皮質感覚野の情報処理を模擬した自己組織化マップが用いられる．本論文の目的は，自己組織化マップを用いて，人の乳児が母国語の音素体系を獲得する過程を計算論的に解析することである．先行研究では自然な入力信号の分布特性（音素の出現頻度やフレーム数）が十分に考慮されていなかった．本研究では，自然な連続音声を入力としてシミュレーション実験を行なった．実験の結果，100秒程度の日本語の連続音声から，5%（/u/）～92%（/s/）の精度で音素体系が獲得可能であることが示された．

書誌： The 1st International Conference on Human-Agent Interaction , 2013

Yoshimasa TAWATSUJI, Keiichi MURAMATSU, Kazuaki KOJIMA, Tatsunori MATSUI: Explanation for Human Sensitive Response to a Humanlike Agent Focusing on Amygdalar Function, The 1st International Conference on Human-Agent Interaction, II-p20 (2013)

Abstract:The uncanny valley is a critical issue in the research eld of Human-Computer Interaction. In the human information processing, an agent highly similar to a real human is rst perceived as a human and subsequently as a non-human. The dual-pathway of emotion can pro-

vide an explanation for this two human processes to the agent, and the inconsistency of processed information, human and non-human, can cause the occurrence of the uncanny valley. We propose a model of how possible the inconsistency occurs, based on the dual-pathway of emotion and the function of the amygdala. According to the model, the inconsistency can the most likely occur where an agent looks highly but imperfectly human.

書誌： 第71回先進的学習科学と工学研究会 , 2014

田和辻可昌, 村松慶一, 松居辰則：定性表現を用いた人型エージェントの目の形態的特徴が人間の情動状態に与える影響の理論的検討, 第71回先進的学習科学と工学研究会, B401-4, pp.17-22 (2014)

Abstract: Designing the agents extraordinarily humanlike sometimes makes humans elicit a negative feeling against it. It is well known as the uncanny valley. We proposed the model generating how the feeling occurs based on human brain function related to emotions. However, the model was defective in its emotional information processing for the inputs, the features of the agents' eye. In this study we propose the advanced model to express the human information

processing toward them. The results of simulation indicated that a little detection of abnormality of the eye in the early but imprecise information processing caused the agent to be low-evaluated, and that agent should be negatively evaluated if human kept on gazing at an abnormality point.

書誌： 総合リハビリテーション ,2014

村岡慶裕、2014、スマートフォンを用いた低コスト筋電図バイオフィードバック装置の開発、 総合リハビリテーション、 Vol.42、No.2、 161-164

☆

対話音声とプロセーシス学、   市川　熹 、プロセーシス学の諸相、森本一誠、神田和幸編著、第1章感覚系しょうがい支援、ユニオンプレス, pp.55-66, (2014.９)

書誌： Medical science digest ,2013

Abstract: 携帯型バイオフィードバック機能付低周波治療器である随意運動介助型電気刺激装置（Integrated Volitionalcontrol Electrical Stimulator: IVES）は，その開発から10年の歳月を経て市販化された。本稿では，希少な国産医療機器であるIVESの持つ様々な機能や効果，応用例などについて述べた。Keywords: : バイオフィードバック，低周波治療器，脳卒中，治療的電気刺激TES

書誌：　 第67回人工知能学会先進的学習科学と工学研究会 ,2013

田和辻可昌, 小島一晃, 松居辰則: 感情の二重経路における脳機能に着目した不気味の谷発生メカニズムのモデル化の試み, 第67回人工知能学会先進的学習科学と工学研究会 (2013)

Abstract: For designing the rounded communication between humans and agents, humanlike appearance of agents can contribute to human understandibility toward its intension. However, the excessive humanlikeness can cause human to feel repulsive against agent, which is well known as the uncanny valley. In this study we propose the model indicating how the human negative response occurs, based on the brain regions and its functions, including the amygdala, cortex and

hippocampus. This model is discribled with quantitative reasoning. The results indicate that as human observes a humanlike agent, the emotional response goes negative and the brain regions were more activated in comparison with the case human observes a person.

書誌： PLoS One ,2013

Ricardo Augusto Hoffmann BION，MIYAZAWA Kouki，KIKUCHI  Hideaki，MAZUKA Reiko，”Learning phonemic vowel length from naturalistic recordings of  Japanese infantdirected speech”，PLoS  One，vol.8，no.2，pp.e51594，2013

Abstract

In Japanese, vowel duration can distinguish the meaning of words. In order for infants to learn this phonemic contrast using simple distributional analyses, there should be reliable differences in the duration of short and long vowels, and the frequency distribution of vowels must make these differences salient enough in the input. In this study, we evaluate these requirements of phonemic learning by analyzing the duration of vowels from over 11 hours of Japanese infant-directed speech. We found that long vowels are substantially longer than short vowels in the input directed to infants, for each of the five oral vowels. However, we also found that learning phonemic length from the overall distribution of vowel duration is not going to be easy for a simple distributional learner, because of the large base-rate effect (i.e., 94% of vowels are short), and because of the many factors that influence vowel duration (e.g., intonational phrase boundaries, word boundaries, and vowel height). Therefore, a successful learner would need to take into account additional factors such as prosodic and lexical cues in order to discover that duration can contrast the meaning of words in Japanese. These findings highlight the importance of taking into account the naturalistic distributions of lexicons and acoustic cues when modeling early phonemic learning.

書誌： Tokai  J Exp Clin Med , 2013

Aono K, Kodama M, Masakado Y, Muraoka Y. (2013). Tokai J Exp Clin Med. 20;38(1):1-6. Objective: Changes in cortical excitability during motor imagery were investigated in order to reveal the effect of hand dominance. During motor imagery, motor evoked potentials (MEPs) were recorded from the first dorsal interosseous (FDI) muscle of the dominant hand using transcranial magnetic stimulation (TMS). Methods: Twelve healthy right-handed subjects participated. Three motor imagery tasks (MITs) were provided; dominant hand grasping, non-dominant hand grasping, and ankle dorsiflexion ipsilateral to the dominant hand. MEPs were also recorded from the FDI muscle of the non-dominant hand during the same tasks. Result: MEPs increased significantly in the dominant hand during MIT, just before MIT of the dominant hand, and prior to ankle dorsiflexion ipsilateral to the dominant hand. MEPs obtained from the FDI muscle of the dominant hand during MITs were greater than that obtained from the FDI muscle of the non-dominant hand. However, this difference was not significant. Conclusion: The left primary motor cortex (M1) was more excited than M1 during MITs of the hand muscles. Cortical excitability increased just before MIT of the contralateral hand and leg muscles. Key words: cortical excitability, motor imagery, transcranial magnetic stimulation, motor evoked potential, Lateralization 

☆

書誌： 日本保健科学学会誌 ,2013

神谷晃央・竹井仁・武田湖太郎・村岡慶裕・笹崎義弘、2013、全人工股関節置換術前の逆トレ ンデレンブルク歩行の有無による前額面における 歩行時姿勢や運動機能と回復過程の差異、日本保健科学学会誌、Vol.15、No.4、219-230 

書誌： 自然 言語処理 ,2014

沈 睿，菊池 英明，"音声言語コーパスにおけるspeaking styleの自動推定 ―転記テキストに着目して―"，自然言語処理，vol.21，no.3，pp.445-464，2014.

要約 近年，計算機技術の進歩に伴って大規模言語データの蓄積と処理が容易となり，音声言語コーパスの構築と実用化の研究が盛んに行われている．我々は，speaking style に関心を持つ利用者に音声言語コーパスを探しやすくさせるために，音声言語コーパスの speaking style の自動推定を目指している．本研究では，Eskenazi が提唱した speaking style の 3 尺度を導入し，従来の文体・ジャンルの判別や著者推定などの自然言語処理の分野で用いられた言語の形態論的特徴を手がかりとし，音声に付随する書き起こしテキスト(本論文では転記テキストと呼ぶ)に着目したspeaking style 推定モデルの構築を試みた．具体的な手続きとしては，はじめに様々な音声言語コーパスから音声に付随する転記テキストを無作為に抽出する．次にこれらの転記テキストを刺激として用い，3尺度のspeaking styleの評定実験を行う．そして，評定結果を目的変数，転記テキストの品詞・語種率と形態素パタンを説明変数とし，重回帰分析により 3 尺度それぞれの回帰モデルを求める．交差検定を行った結果，本研究の提案手法によって 3 尺度の内 2 尺度の speaking style 評定値を高い精度で推定できることを確認した．