A computational passage-of-time model of the cerebellar Purkinje cell in eyeblink conditioning.

The cerebellar Purkinje cell controlling eyeblinks can learn, remember, and reproduce the interstimulus interval in a classical conditioning paradigm. Given temporally separated inputs, the cerebellar Purkinje cell learns to pause its tonic inhibition of a motor pathway with high temporal precision so that an overt blink occurs at the right time. Most models place the passage-of-time representation in upstream network effects. Yet, bypassing the upstream network and directly stimulating the Purkinje cell's pre-synaptic fibers during conditioning still causes acquisition of a well-timed response. Additionally, while network models are sensitive to variance in the temporal structure of probe stimulation, in vivo findings suggest that the acquired Purkinje cell response is not. Such findings motivate alternative approaches to modeling neural function. Here, we present a proof-of-principle model of the passage-of-time which is internal to the Purkinje cell and is invariant to probe structure. The model is consistent with puzzling findings, accurately recapitulates Purkinje cell firing during classical conditioning and makes testable electrophysiological predictions.

Differential Involvement of EEG Oscillatory Components in Sameness versus Spatial-Relation Visual Reasoning Tasks.

The development of deep convolutional neural networks (CNNs) has recently led to great successes in computer vision, and CNNs have become de facto computational models of vision. However, a growing body of work suggests that they exhibit critical limitations on tasks beyond image categorization. Here, we study one such fundamental limitation, concerning the judgment of whether two simultaneously presented items are the same or different (SD) compared with a baseline assessment of their spatial relationship (SR). In both human subjects and artificial neural networks, we test the prediction that SD tasks recruit additional cortical mechanisms which underlie critical aspects of visual cognition that are not explained by current computational models. We thus recorded electroencephalography (EEG) signals from human participants engaged in the same tasks as the computational models. Importantly, in humans the two tasks were matched in terms of difficulty by an adaptive psychometric procedure; yet, on top of a modulation of evoked potentials (EPs), our results revealed higher activity in the low ß (16-24 Hz) band in the SD compared with the SR conditions. We surmise that these oscillations reflect the crucial involvement of additional mechanisms, such as working memory and attention, which are missing in current feed-forward CNNs.

Not-So-CLEVR: learning same-different relations strains feedforward neural networks.

The advent of deep learning has recently led to great successes in various engineering applications. As a prime example, convolutional neural networks, a type of feedforward neural network, now approach human accuracy on visual recognition tasks like image classification and face recognition. However, here we will show that feedforward neural networks struggle to learn abstract visual relations that are effortlessly recognized by non-human primates, birds, rodents and even insects. We systematically study the ability of feedforward neural networks to learn to recognize a variety of visual relations and demonstrate that same-different visual relations pose a particular strain on these networks. Networks fail to learn same-different visual relations when stimulus variability makes rote memorization difficult. Further, we show that learning same-different problems becomes trivial for a feedforward network that is fed with perceptually grouped stimuli. This demonstration and the comparative success of biological vision in learning visual relations suggests that feedback mechanisms such as attention, working memory and perceptual grouping may be the key components underlying human-level abstract visual reasoning.

A systematic comparison between visual cues for boundary detection.

The detection of object boundaries is a critical first step for many visual processing tasks. Multiple cues (we consider luminance, color, motion and binocular disparity) available in the early visual system may signal object boundaries but little is known about their relative diagnosticity and how to optimally combine them for boundary detection. This study thus aims at understanding how early visual processes inform boundary detection in natural scenes. We collected color binocular video sequences of natural scenes to construct a video database. Each scene was annotated with two full sets of ground-truth contours (one set limited to object boundaries and another set which included all edges). We implemented an integrated computational model of early vision that spans all considered cues, and then assessed their diagnosticity by training machine learning classifiers on individual channels. Color and luminance were found to be most diagnostic while stereo and motion were least. Combining all cues yielded a significant improvement in accuracy beyond that of any cue in isolation. Furthermore, the accuracy of individual cues was found to be a poor predictor of their unique contribution for the combination. This result suggested a complex interaction between cues, which we further quantified using regularization techniques. Our systematic assessment of the accuracy of early vision models for boundary detection together with the resulting annotated video dataset should provide a useful benchmark towards the development of higher-level models of visual processing.

Preparation of transparent conductive multilayered films using active pentafluorophenyl ester modified multiwalled carbon nanotubes.

A multilayered film was prepared by layer-by-layer (LBL) assembly of active ester modified multiwalled carbon nanotubes (MWCNTs) and poly(allylamine hydrochloride) (PAH). For this purpose, carboxylic groups on the surface of the oxidized MWCNTs were converted to the acyl chlorides by their reaction with thionyl chloride. Subsequent reaction of the acyl chlorides with pentafluorophenol formed the active esters. These active ester modified MWCNTs (MWCNTs-COOC(6)F(5)) were air-stable and moisture resistant, but showed a high reactivity toward primary or secondary amines resulting in amide bonds. For the preparation of a multilayered film, the surface of a quartz slide was first activated and sacrificial double layers of PAH and poly(sodium 4-styrene sulfonate) (PSS) were deposited. Subsequently, LBL assembly of MWCNTs-COOC(6)F(5) and PAH was then conducted on these double layers [(PAH/PSS)2]. In the process of the assembly, a reaction occurred between the active ester on the surface of MWCNTs and the amine groups of polyallylamine yielding amide bonds, which resulted in a mechanically stable thin film. A free-standing film was obtained after dissolving the sacrificial layer [(PAH/PSS)2] in a concentrated aqueous NaOH solution. The surface resistance of the multilayered film with 20 bilayers decreased to around 10 kOmega while remaining a reasonable transparency (70% at 500 nm).

Mechanistic examination of pre-exfoliating confinement of surface-active polystyrene nanobeads within pristine clay.

Hydrophobic polymer (PS) nanoparticles preformed through an emulsifier-free emulsion polymerization method were successfully incorporated into a gallery of pristine sodium montmorillonite via interfacial cation exchange. The polymer beads confined between clay nanosheets were capable of (1) preventing the silicate layers from restacking and (2) maintaining the exfoliated state of clay. The increase in the abundance of surface groups promoted adsorption of the nanobeads onto the silicate surface and eventually led to the establishment of strong polymer-clay interactions. These findings suggest that, on the basis of the obtained pre-exfoliated clay masterbatch, the presence of strong polymer-clay interactions could improve the mechanical performance of nanocomposites.