Deep learning and punctuated equilibrium theory

Deep learning is associated with the latest success stories in AI. In particular, deep neural networks are applied in increasingly different fields to model complex processes. Interestingly, the underlying algorithm of backpropagation was originally designed for political science models. The theoretical foundations of this approach are very similar to the concept of Punctuated Equilibrium Theory (PET). The article discusses the concept of deep learning and shows parallels to PET. A showcase model demonstrates how deep learning can be used to provide a missing link in the study of the policy process: the connection between attention in the political system (as inputs) and budget shifts (as outputs).     

Comparing different kinds of words and word-word relations to test an habituation model of priming

Huber and O'Reilly (2003) proposed that neural habituation exists to solve a temporal parsing problem, minimizing blending between one word and the next when words are visually presented in rapid succession. They developed a neural dynamics habituation model, explaining the finding that short duration primes produce positive priming whereas long duration primes produce negative repetition priming. The model contains three layers of processing, including a visual input layer, an orthographic layer, and a lexical-semantic layer. The predicted effect of prime duration depends both on this assumed representational hierarchy and the assumption that synaptic depression underlies habituation. The current study tested these assumptions by comparing different kinds of words (e.g., words versus non-words) and different kinds of word-word relations (e.g., associative versus repetition). For each experiment, the predictions of the original model were compared to an alternative model with different representational assumptions. Experiment 1 confirmed the prediction that non-words and inverted words require longer prime durations to eliminate positive repetition priming (i.e., a slower transition from positive to negative priming). Experiment 2 confirmed the prediction that associative priming increases and then decreases with increasing prime duration, but remains positive even with long duration primes. Experiment 3 replicated the effects of repetition and associative priming using a within-subjects design and combined these effects by examining target words that were expected to repeat (e.g., viewing the target word ‘BACK' after the prime phrase ‘back to'). These results support the originally assumed representational hierarchy and more generally the role of habituation in temporal parsing and priming.     

发展性协调障碍的神经机制

发展性协调障碍是一种特殊的发育障碍, 它的一个显著特征是动作协调困难, 这种障碍影响着5%~8%学龄儿童的学业成绩和日常生活。发展性协调障碍常常与其他类型的学习困难并存。发展性协调障碍的病因非常复杂, 这种障碍与一定程度的视觉空间认知能力受损有关, 早期精细运动技能发育与脑神经发育进程可能存在时间的重合。近年来, 研究者们试图探索影响发展性协调障碍的神经基础, 并先后提出不同的神经关联假说。这些理论假说主要涉及到小脑、顶叶、胼胝体、基底神经节和大脑白质等脑区。本文全面回顾了该领域研究的最新进展, 并提出未来研究应该关注的方向与问题。     

Neural circuitry of the bilingual mental lexicon: Effect of age of second language acquisition

Numerous studies have proposed that changes of the human language faculty caused by neural maturation can explain the substantial differences in ultimate attainment of grammatical competences between first language (L1) acquirers and second language (L2) learners. However, little evidence on the effect of neural maturation on the attainment of lexical knowledge in L2 is available. The present functional magnetic resonance study addresses this question via a cross-linguistic neural adaptation paradigm. Age of acquisition (AoA) of L2 was systematically manipulated. Concrete nouns were repeated across language (e.g., French–German, valise^suitcase–Koffer^suitcase). Whereas early bilinguals (AoA of L2 < 3 years) showed larger repetition enhancement (RE) effects in the left superior temporal gyrus, the bilateral superior frontal gyrus and the right posterior insula, late bilinguals (AoA of L2 > 10 years) showed larger RE effects in the middle portion of the left insula and in the right middle frontal gyrus (MFG). We suggest that, as for grammatical knowledge, the attainment of lexical knowledge in L2 is affected by neural maturation. The present findings lend support to neurocognitive models of bilingual word recognition postulating that, for both early and late bilinguals, the two languages are interconnected at the conceptual level.     

The duality principle: irreducibility of sub-threshold psychophysical computation to neuronal brain activation

A key working hypothesis in neuroscience is ‘materialistic reductionism’, i.e., the assumption whereby all physiological, behavioral or cognitive phenomena is produced by localized neurochemical brain activation (but not vice versa). However, analysis of sub-threshold Weber’s psychophysical stimulation indicates its computational irreducibility to the direct interaction between psychophysical stimulation and any neuron/s. This is because the materialistic-reductionistic working hypothesis assumes that the determination of the existence or non-existence of any psychophysical stimulation [s] may only be determined through its direct interaction [di1] with a given neuron/s [N] that together forms the ‘neural registry’ computational level [NR/di1]. But, this implies that in cases of (initial) sub-threshold (sensory-specific) psychophysical stimulation which is increased above the sensory-specific threshold but below Weber’s psychophysical ‘dv’—the psychophysical computational processing [PCP] produces an apparently ‘computationally indeterminate’ output. This is because materialistic reductionism asserts the contingency of PCP upon the existence of a direct interaction between ‘s’ and ‘N’ within the NR/di1 level, but in the special case of Weber’s sub-threshold psychophysical stimulation the same PCP/di1 also asserts the non-existence of ‘s’ (as demanded by Weber’s psychophysical law). However, given robust empirical evidence indicating the capability of PCP to determine whether (or not) ‘s’ exists, we must conclude that PCP may not be carried out from within NR’s direct interaction between a particular psychophysical stimulation and any set of neuron/s in the brain. Hence, the Duality Principle asserts the conceptual irreducibility of sub-threshold psychophysical stimulation to any direct NR/di1: s-N interaction, thereby challenging the current materialistic-reductionistic assumption.     

Doing cognitive neuroscience: a third way

The “top-down” and “bottom-up” approaches have been thought to exhaust the possibilities for doing cognitive neuroscience. We argue that neither approach is likely to succeed in providing a theory that enables us to understand how cognition is achieved in biological creatures like ourselves. We consider a promising third way of doing cognitive neuroscience, what might be called the “neural dynamic systems” approach, that construes cognitive neuroscience as an autonomous explanatory endeavor, aiming to characterize in its own terms the states and processes responsible for brain-based cognition. We sketch the basic motivation for the approach, describe a particular version of the approach, so-called ‘Dynamic Causal Modeling’ (DCM), and consider a concrete example of DCM. This third way, we argue, has the potential to avoid the problems that afflict the other two approaches.     

Application of improved BP neural network based on e-commerce supply chain network data in the forecast of aquatic product export volume

Aiming at the existing problems in the production and export scale prediction of aquaculture, a model of yield prediction based on BP Neural network algorithm is proposed, and a set of algorithms is proposed to optimize BP neural network (BPNN). Based on the traditional BP neural network, it is easy to get into the local optimal problem due to the long training time of the model. By using the simple Johnson algorithm, the dimensionality of the input neuron is reduced, and then the hidden layer neural network is determined by this method. At the same time, the data mining method is used to filter the Data.Particle swarm optimization algorithm is used to optimize the parameters. At the same time, based on the domestic e-commerce Sales network data, the results show that the average square root error of the model is less than the traditional BP neural network and the learning efficiency is higher than the traditional BP neural network. The results show that the model has a great advantage in building up a large number of historical data, and it can shorten the modeling time and get good prediction result by combining the sales data of e-commerce. It provides a new feasible method for the export prediction of aquatic products.     

The role of the magnocellular and parvocellular pathways in the attentional blink

The attentional blink refers to the transient impairment in perceiving the 2nd of two targets presented in close temporal proximity in a rapid serial visual presentation (RSVP) stream. The purpose of this study was to examine the effect on human attentional-blink performance of disrupting the function of the magnocellular pathway--a major visual-processing pathway specialized in temporal segregation. The study was motivated by recent theories that relate the attentional blink to the limited temporal resolution of attentional responses, and by a number of poorly understood empirical findings, including the effects on the attentional blink of luminance adaptation and distraction. The attentional blink was assessed for stimuli on a red background (Experiment 1), stimuli on an equiluminant background (Experiment 2), and following flicker or motion adaptation (Experiment 3), three psychophysical manipulations known to disrupt magnocellular function. Contrary to our expectations, the attentional blink was not affected by these manipulations, suggesting no specific relationship between the attentional blink and magnocellular and/or parvocellular processing.     

Interactive computation is interaction with what?: A reply to Clark

In this response essay, we argue that Andy Clark’s assessment of our position on cyborgs is rooted in a misconception of the notion of “interaction” that we advance.     

Finding ERP-signatures of target awareness: Puzzle persists because of experimental co-variation of the objective and subjective variables

Using masking techniques combined with electrophysiological recordings is a promising way to study neural correlates of visual awareness, as shown in recent studies. Here I comment on the following puzzling aspects typical for this endeavour that have made obstacles for a potentially even more impressive progress. First, the continuing practice of confounds between objective stimulus variables and subjective dependent measures. Second, complexity of timing the emergence of subjective conscious percept which is partly due to complex interactivity between target and mask. Third, the ambiguity of the concept of neural correlates of awareness.