ATR Computational Neuroscience Laboratories
ATR Computational Neuroscience Laboratories
Tong F.,Vanderbilt University |
Harrison S.A.,Vanderbilt University |
Dewey J.A.,Vanderbilt University |
Kamitani Y.,ATR Computational Neuroscience Laboratories
NeuroImage | Year: 2012
Orientation-selective responses can be decoded from fMRI activity patterns in the human visual cortex, using multivariate pattern analysis (MVPA). To what extent do these feature-selective activity patterns depend on the strength and quality of the sensory input, and might the reliability of these activity patterns be predicted by the gross amplitude of the stimulus-driven BOLD response? Observers viewed oriented gratings that varied in luminance contrast (4, 20 or 100%) or spatial frequency (0.25, 1.0 or 4.0. cpd). As predicted, activity patterns in early visual areas led to better discrimination of orientations presented at high than low contrast, with greater effects of contrast found in area V1 than in V3. A second experiment revealed generally better decoding of orientations at low or moderate as compared to high spatial frequencies. Interestingly however, V1 exhibited a relative advantage at discriminating high spatial frequency orientations, consistent with the finer scale of representation in the primary visual cortex. In both experiments, the reliability of these orientation-selective activity patterns was well predicted by the average BOLD amplitude in each region of interest, as indicated by correlation analyses, as well as decoding applied to a simple model of voxel responses to simulated orientation columns. Moreover, individual differences in decoding accuracy could be predicted by the signal-to-noise ratio of an individual's BOLD response. Our results indicate that decoding accuracy can be well predicted by incorporating the amplitude of the BOLD response into simple simulation models of cortical selectivity; such models could prove useful in future applications of fMRI pattern classification. © 2012 Elsevier Inc.
Horikawa T.,ATR Computational Neuroscience Laboratories |
Kamitani Y.,Kyoto University
Nature Communications | Year: 2017
Object recognition is a key function in both human and machine vision. While brain decoding of seen and imagined objects has been achieved, the prediction is limited to training examples. We present a decoding approach for arbitrary objects using the machine vision principle that an object category is represented by a set of features rendered invariant through hierarchical processing. We show that visual features, including those derived from a deep convolutional neural network, can be predicted from fMRI patterns, and that greater accuracy is achieved for low-/high-level features with lower-/higher-level visual areas, respectively. Predicted features are used to identify seen/imagined object categories (extending beyond decoder training) from a set of computed features for numerous object images. Furthermore, decoding of imagined objects reveals progressive recruitment of higher-to-lower visual representations. Our results demonstrate a homology between human and machine vision and its utility for brain-based information retrieval. © The Author(s) 2017.
Lisi G.,ATR Computational Neuroscience Laboratories |
Lisi G.,Nara Institute of Science and Technology |
Noda T.,ATR Computational Neuroscience Laboratories |
Morimoto J.,ATR Computational Neuroscience Laboratories
Frontiers in Systems Neuroscience | Year: 2014
This paper investigates the influence of the leg afferent input, induced by a leg assistive robot, on the decoding performance of a BMI system. Specifically, it focuses on a decoder based on the event-related (de)synchronization (ERD/ERS) of the sensorimotor area. The EEG experiment, performed with healthy subjects, is structured as a 3 × 2 factorial design, consisting of two factors: "finger tapping task" and "leg condition." The former is divided into three levels (BMI classes), being left hand finger tapping, right hand finger tapping and no movement (Idle); while the latter is composed by two levels: leg perturbed (Pert) and leg not perturbed (NoPert). Specifically, the subjects' leg was periodically perturbed by an assistive robot in 5 out of 10 sessions of the experiment and not moved in the remaining sessions. The aim of this study is to verify that the decoding performance of the finger tapping task is comparable between the two conditions NoPert and Pert. Accordingly, a classifier is trained to output the class of the finger tapping, given as input the features associated with the ERD/ERS. Individually for each subject, the decoding performance is statistically compared between the NoPert and Pert conditions. Results show that the decoding performance is notably above chance, for all the subjects, under both conditions. Moreover, the statistical comparison do not highlight a significant difference between NoPert and Pert in any subject, which is confirmed by feature visualization. © 2014 Lisi, Noda and Morimoto.
Callan D.,Atr Computational Neuroscience Laboratories |
Callan A.,Nict Multimodal Communication Group |
Gamez M.,Atr Computational Neuroscience Laboratories |
Sato M.-A.,Atr Computational Neuroscience Laboratories |
Kawato M.,Atr Computational Neuroscience Laboratories
NeuroImage | Year: 2010
Articulatory goals have long been proposed to mediate perception. Examples include direct realist and constructivist (analysis by synthesis) theories of speech perception. Although the activity in brain regions involved with action production has been shown to be present during action observation (Mirror Neuron System), the relationship of this activity to perceptual performance has not been clearly demonstrated at the event level. To this end we used functional magnetic resonance imaging fMRI and magnetoencephalography MEG to measure brain activity for correct and incorrect trials of an auditory phonetic identification in noise task. FMRI analysis revealed activity in the premotor cortex including the neighboring frontal opercular part of Broca's area (PMC/Broca's) for both perception and production tasks involving the same phonetic stimuli (potential mirror system site) that was significantly greater for correct over incorrect perceptual identification trials. Time-frequency analysis of single trials conducted over MEG current localized to PMC/Broca's using a hierarchical variational Bayesian source analysis technique revealed significantly greater event-related synchronization ERS and desynchronization ERD for correct over incorrect trials in the alpha, beta, and gamma frequency range prior to and after stimulus presentation. Together, these fMRI and MEG results are consistent with the hypothesis that articulatory processes serve to facilitate perceptual performance, while further dispelling concerns that activity found in ventral PMC/Broca's (mirror system) is merely a product of covert production of the perceived action. The finding of performance predictive activity prior to stimulus onset as well as activity related to task difficulty instead of information available in stimulation are consistent with constructivist and contrary to direct realist theories of perception. © 2010 Elsevier Inc.
Kawato M.,ATR Computational Neuroscience Laboratories |
Kuroda S.,University of Tokyo |
Schweighofer N.,University of Southern California
Current Opinion in Neurobiology | Year: 2011
The biophysical models of spike-timing-dependent plasticity have explored dynamics with molecular basis for such computational concepts as coincidence detection, synaptic eligibility trace, and Hebbian learning. They overall support different learning algorithms in different brain areas, especially supervised learning in the cerebellum. Because a single spine is physically very small, chemical reactions at it are essentially stochastic, and thus sensitivity-longevity dilemma exists in the synaptic memory. Here, the cascade of excitable and bistable dynamics is proposed to overcome this difficulty. All kinds of learning algorithms in different brain regions confront with difficult generalization problems. For resolution of this issue, the control of the degrees-of-freedom can be realized by changing synchronicity of neural firing. Especially, for cerebellar supervised learning, the triangle closed-loop circuit consisting of Purkinje cells, the inferior olive nucleus, and the cerebellar nucleus is proposed as a circuit to optimally control synchronous firing and degrees-of-freedom in learning. © 2011 Elsevier Ltd.
Sasaki Y.,Massachusetts General Hospital |
Sasaki Y.,Harvard University |
Sasaki Y.,ATR Computational Neuroscience Laboratories |
Nanez J.E.,Arizona State University |
And 2 more authors.
Nature Reviews Neuroscience | Year: 2010
Visual perceptual learning (VPL) is defined as a long-term improvement in performance on a visual task. In recent years, the idea that conscious effort is necessary for VPL to occur has been challenged by research suggesting the involvement of more implicit processing mechanisms, such as reinforcement-driven processing and consolidation. In addition, we have learnt much about the neural substrates of VPL and it has become evident that changes in visual areas and regions beyond the visual cortex can take place during VPL. © 2010 Macmillan Publishers Limited. All rights reserved.
Kamitani Y.,ATR Computational Neuroscience Laboratories |
Sawahata Y.,ATR Computational Neuroscience Laboratories
NeuroImage | Year: 2010
Op de Beeck (Op de Beeck, H., 2009. Against hyperacuity in brain reading: Spatial smoothing does not hurt multivariate fMRI analyses? Neuroimage) challenges the possibility of extracting information from subvoxel representations via random biases associated with voxel sampling, the hypothesis proposed by Kamitani and Tong (Kamitani, Y., Tong, F., 2005. Decoding the visual and subjective contents of the human brain. Nat. Neurosci. 8, 679-685). Here, we show that his results provide no evidence against the possibility, being consistent with both of the subvoxel and supravoxel representation models. Classification of spatially smoothed fMRI data is not an effective means to probe into information sources for multivoxel decoding, since smoothing does not hurt the information contents of multivoxel patterns. We point out the danger of interpreting multivoxel decoding results based on intuitions guided by the conventional brain mapping paradigm. © 2009 Elsevier Inc. All rights reserved.
Matsubara T.,Nara Institute of Science and Technology |
Matsubara T.,ATR Computational Neuroscience Laboratories |
Morimoto J.,ATR Computational Neuroscience Laboratories
IEEE Transactions on Biomedical Engineering | Year: 2013
In this study, we propose a multiuser myoelectric interface that can easily adapt to novel users. When a user performs different motions (e.g., grasping and pinching), different electromyography (EMG) signals are measured. When different users perform the same motion (e.g., grasping), different EMG signals are also measured. Therefore, designing a myoelectric interface that can be used by multiple users to perform multiple motions is difficult. To cope with this problem, we propose for EMG signals a bilinear model that is composed of two linear factors: 1) user dependent and 2) motion dependent. By decomposing the EMG signals into these two factors, the extracted motion-dependent factors can be used as user-independent features. We can construct a motion classifier on the extracted feature space to develop the multiuser interface. For novel users, the proposed adaptation method estimates the user-dependent factor through only a few interactions. The bilinear EMG model with the estimated user-dependent factor can extract the user-independent features from the novel user data. We applied our proposed method to a recognition task of five hand gestures for robotic hand control using four-channel EMG signals measured from subject forearms. Our method resulted in 73% accuracy, which was statistically significantly different from the accuracy of standard nonmultiuser interfaces, as the result of a two-sample t -test at a significance level of 1%. © 1964-2012 IEEE.
News Article | March 29, 2016
People with bipolar disorder, also known as manic-depressive disorder, experience extreme fluctuations in mood and behavior, which may occur in cycles lasting for days, months, or years. Such changes were first described more than half a century ago, but their molecular basis in the brain has remained unclear. Now, researchers at the Institute for Comprehensive Medical Science, Fujita Health University, and ATR Computational Neuroscience Laboratories in Japan have succeeded in predicting states of mood-change-like behavior by studying the gene expression patterns in the brain in a bipolar disorder mouse model. Interestingly, so-called circadian genes, whose expressions increase and decrease over a 24-hour cycle, are overrepresented in the prediction gene sets, demonstrating an intrinsic link between circadian genes in the brain and mood change-like behavior. The research appears in the March 29th the journal Cell Reports. Elucidation of the molecular basis of mood changes occurring with an infradian (longer than a day) rhythm has been hampered by the lack of an animal model that exhibits spontaneous behavioral changes related to the infradian oscillation of mood. In the course of screening over 180 mutant mouse strains with a systematic battery of behavioral tests, Dr. Tsuyoshi Miyakawa and his colleagues found that mice with heterozygous knockout of the alpha-isoform of calcium/calmodulin-dependent protein kinase II (αCaMKII) exhibit behavioral deficits and other brain features consistent with bipolar disorder. Notably, the mutant mice also showed periodic changes in locomotor activity in their home cages with an approximate cycle length of 10-20 days. The changes in locomotor activity are associated with fluctuations of anxiety-like and depression-like behaviors, suggesting that the mutant mice may serve as an animal model showing infradian oscillations of mood substantially similar to those found in patients with bipolar disorder. A recent human study also indicated a genetic association of the αCaMKII gene with bipolar disorder, and decreased expression of αCaMKII has been observed in postmortem brains of patients with bipolar disorder. These findings indicate that αCaMKII mutant mice should serve as a good animal model of bipolar disorder, to elucidate the pathogenesis and pathophysiology of the disorder. In the current study, the researchers used infradian cyclic locomotor activity in the mutant mice as a proxy for mood-associated changes, and examined their molecular basis in the brain by conducting prediction analyses of the gene expression data. At first, researchers longitudinally monitored locomotor activity of 37 αCaMKII mutant mice by calculating the distance traveled in their cage for over 2 months. Subsequently, researchers dissected the hippocampus, a region thought to be involved in the regulation of mood, from the brain. The sampling of the hippocampus was conducted at the same time of the day (between 1 and 2 p.m.). Gene expression patterns in the hippocampus samples were examined using DNA microarrays that measured the expression levels of over 30,000 genes (transcripts) per sample. Based on the gene expression data, they constructed models for retrospectively predicting locomotor activity of individual mice. The researchers found that gene expression patterns in the hippocampus accurately predicted whether the mice were in a state of high or low locomotor activity. "This is the first demonstration, to our knowledge, of successful quantitative predictions of the individual behavioral state from gene expression patterns in the brain of a mammal," says Miyakawa. "Gene expression patterns in the hippocampus may retain information about past locomotor activity." In the current study, prediction analysis of gene expression data was implemented in order to identify the genes that are most useful to determine the state of cyclic changes in locomotor activity. Thus, the researchers examined the list of genes used for the successful prediction of locomotor activity. "To our surprise, the list of 'prediction genes' included significantly higher number of circadian genes, genes that are known to fluctuate according to circadian rhythms. Circadian genes turned out to be also infradian genes, whose expressions go up and down with mood-change-like behaviors in these mice," Miyakawa explains. Researchers also found that levels of cAMP and pCREB, possible upstream regulators of some circadian genes, were correlated with locomotor activity. "The current results provide the evidence for a novel concept that some circadian genes and their regulatory machinery in the brain may be involved in the generation of infradian rhythm behavior," Miyakawa explains. Furthermore, researchers found that drugs that are used to treat bipolar disorder controlled the locomotor activity and changed hippocampal pCREB expression in the mutant mice. These results support the idea that hippocampal pCREB levels may modulate locomotor activity. "While the work so far has been limited to a mouse model of bipolar disorder, regulating effectively such molecular changes might lead to treatment for the disorder," Miyakawa says. "It is also of interest whether certain molecular signatures in the samples, such as blood and cerebrospinal fluid, obtained from living animals can predict past and future locomotor activity. If the successful predictions are confirmed in the mouse model, this strategy may have potential for developing new methods for diagnosis, as well as treatment, of patients with bipolar disorder," Miyakawa says. Explore further: Immaturity of the brain may cause schizophrenia
Choi K.,ATR Computational Neuroscience Laboratories
European Journal of Applied Physiology | Year: 2012
To construct and evaluate a novel wheelchair system that can be freely controlled via electroencephalogram signals in order to allow people paralyzed from the neck down to interact with society more freely. A brain-machine interface (BMI) wheelchair control system was constructed by effective signal processing methods, and subjects were trained by a feedback method to decrease the training time and improve accuracy. The implemented system was evaluated through experiments on controlling bars and avoiding obstacles using three subjects. Furthermore, the effectiveness of the feedback training method was evaluated by comparison with an imaginary movement experiment without any visual feedback for two additional subjects. In the bar-controlling experiment, two subjects achieved a 95.00% success rate, and the third had a 91.66% success rate. In the obstacle avoidance experiment, all three achieved success rate over 90% success rate, and required almost the same amount of time to reach as that when driving with a joystick. In the experiment on imaginary movement without visual feedback, the two additional subjects adapted to the experiment far slower than they did with visual feedback. In this study, the feedback training method allowed subjects to easily and rapidly gain accurate control over the implemented wheelchair system. These results show the importance of the feedback training method using neuroplasticity in BMI systems. © 2011 Springer-Verlag.