Tsung-Ren Huang

ARTSCENE: A neural system for natural scene classification

How do humans rapidly recognize a scene? How can neural models capture this biological competence to achieve state-of-the-art scene classification? The ARTSCENE neural system classifies natural scene photographs by using multiple spatial scales to efficiently accumulate evidence for gist and texture. ARTSCENE embodies a coarse-to-fine Texture Size Ranking Principle whereby spatial attention processes multiple scales of scenic information, from global gist to local textures, to learn and recognize scenic properties. The model can incrementally learn and rapidly predict scene identity by gist information alone, and then accumulate learned evidence from scenic textures to refine this hypothesis. The model shows how texture-fitting allocations of spatial attention, called attentional shrouds, can facilitate scene recognition, particularly when they include a border of adjacent textures. Using grid gist plus three shroud textures on a benchmark photograph dataset, ARTSCENE discriminates 4 landscape scene categories (coast, forest, mountain, and countryside) with up to 91.85% correct on a test set, outperforms alternative models in the literature which use biologically implausible computations, and outperforms component systems that use either gist or texture information alone.

Cortical Dynamics of Contextually Cued Attentive Visual Learning and Search: Spatial and Object Evidence Accumulation

How do humans use target-predictive contextual information to facilitate visual search? How are consistently paired scenic objects and positions learned and used to more efficiently guide search in familiar scenes? For example, humans can learn that a certain combination of objects may define a context for a kitchen and trigger a more efficient search for a typical object, such as a sink, in that context. The ARTSCENE Search model is developed to illustrate the neural mechanisms of such memory-based context learning and guidance and to explain challenging behavioral data on positive-negative, spatial-object, and local-distant cueing effects during visual search, as well as related neuroanatomical, neurophysiological, and neuroimaging data. The model proposes how global scene layout at a first glance rapidly forms a hypothesis about the target location. This hypothesis is then incrementally refined as a scene is scanned with saccadic eye movements. The model simulates the interactive dynamics of object and spatial contextual cueing and attention in the cortical What and Where streams starting from early visual areas through medial temporal lobe to prefrontal cortex. After learning, model dorsolateral prefrontal cortex (area 46) primes possible target locations in posterior parietal cortex based on goal-modulated percepts of spatial scene gist that are represented in parahippocampal cortex. Model ventral prefrontal cortex (area 47/12) primes possible target identities in inferior temporal cortex based on the history of viewed objects represented in perirhinal cortex.

Note on the Zolotarev optimal rational approximation for the overlap Dirac operator

We discuss the salient features of the Zolotarev optimal rational approximation for the inverse square root function, in particular, for its applications in lattice QCD with the overlap Dirac quark. The theoretical error bound for the matrix-vector multiplication

Enhanced Spontaneous Oscillations in the Supplementary Motor Area Are Associated with Sleep-Dependent Offline Learning of Finger-Tapping Motor-Sequence Task

{H}_{w}{(H}_{w}^{2}{)}^{\ensuremath{-}1/2}Y

Task attention facilitates learning of task-irrelevant stimuli.

is derived. We check that the error bound is always satisfied amply, for any QCD gauge configurations we have tested. An empirical formula for the error bound is determined, together with its numerical values (by evaluating elliptic functions) listed in Table II as well as plotted in Fig. 3. Our results suggest that, with the Zolotarev approximation to

Assembling old tricks for new tasks: A neural model of instructional learning and control

{(H}_{w}^{2}{)}^{\ensuremath{-}1/2},

Strategic cognitive sequencing: a computational cognitive neuroscience approach

one can essentially preserve the exact chiral symmetry of the overlap Dirac operator to very high precision, for any gauge configuration on a finite lattice.

Developing the Own-Race Advantage in 4-, 6-, and 9-Month-Old Taiwanese Infants: A Perceptual Learning Perspective

Sleep is beneficial for various types of learning and memory, including a finger-tapping motor-sequence task. However, methodological issues hinder clarification of the crucial cortical regions for sleep-dependent consolidation in motor-sequence learning. Here, to investigate the core cortical region for sleep-dependent consolidation of finger-tapping motor-sequence learning, while human subjects were asleep, we measured spontaneous cortical oscillations by magnetoencephalography together with polysomnography, and source-localized the origins of oscillations using individual anatomical brain information from MRI. First, we confirmed that performance of the task at a retest session after sleep significantly increased compared with performance at the training session before sleep. Second, spontaneous δ and fast-σ oscillations significantly increased in the supplementary motor area (SMA) during post-training compared with pretraining sleep, showing significant and high correlation with the performance increase. Third, the increased spontaneous oscillations in the SMA correlated with performance improvement were specific to slow-wave sleep. We also found that correlations of δ oscillation between the SMA and the prefrontal and between the SMA and the parietal regions tended to decrease after training. These results suggest that a core brain region for sleep-dependent consolidation of the finger-tapping motor-sequence learning resides in the SMA contralateral to the trained hand and is mediated by spontaneous δ and fast-σ oscillations, especially during slow-wave sleep. The consolidation may arise along with possible reorganization of a larger-scale cortical network that involves the SMA and cortical regions outside the motor regions, including prefrontal and parietal regions.

Multistability of the Brain Network for Self-other Processing

Attention plays a fundamental role in visual learning and memory. One highly established principle of visual attention is that the harder a central task is, the more attentional resources are used to perform the task and the smaller amount of attention is allocated to peripheral processing because of limited attention capacity. Here we show that this principle holds true in a dual-task setting but not in a paradigm of task-irrelevant perceptual learning. In Experiment 1, eight participants were asked to identify either bright or dim number targets at the screen center and to remember concurrently presented scene backgrounds. Their recognition performances for scenes paired with dim/hard targets were worse than those for scenes paired with bright/easy targets. In Experiment 2, eight participants were asked to identify either bright or dim letter targets at the screen center while a task-irrelevant coherent motion was concurrently presented in the background. After five days of training on letter identification, participants improved their motion sensitivity to the direction paired with hard/dim targets improved but not to the direction paired with easy/bright targets. Taken together, these results suggest that task-irrelevant stimuli are not subject to the attentional control mechanisms that task-relevant stimuli abide.

A computational system for lattice QCD with overlap Dirac quarks

We can learn from the wisdom of others to maximize success. However, it is unclear how humans take advice to flexibly adapt behavior. On the basis of data from neuroanatomy, neurophysiology, and neuroimaging, a biologically plausible model is developed to illustrate the neural mechanisms of learning from instructions. The model consists of two complementary learning pathways. The slow-learning parietal pathway carries out simple or habitual stimulus–response (S-R) mappings, whereas the fast-learning hippocampal pathway implements novel S-R rules. Specifically, the hippocampus can rapidly encode arbitrary S-R associations, and stimulus-cued responses are later recalled into the basal ganglia-gated pFC to bias response selection in the premotor and motor cortices. The interactions between the two model learning pathways explain how instructions can override habits and how automaticity can be achieved through motor consolidation.