The Hebbian paradigm reintegrated: Local reverberations as internal representations

External Stimuli on Neural Networks: Analytical and Numerical Approaches

Based on the behavior of living beings, which react mostly to external stimuli, we introduce a neural-network model that uses external patterns as a fundamental tool for the process of recognition. In this proposal, external stimuli appear as an additional field, and basins of attraction, representing memories, arise in accordance with this new field. This is in contrast to the more-common attractor neural networks, where memories are attractors inside well-defined basins of attraction. We show that this procedure considerably increases the storage capabilities of the neural network; this property is illustrated by the standard Hopfield model, which reveals that the recognition capacity of our model may be enlarged, typically, by a factor 102. The primary challenge here consists in calibrating the influence of the external stimulus, in order to attenuate the noise generated by memories that are not correlated with the external pattern. The system is analyzed primarily through numerical simulations. However, since there is the possibility of performing analytical calculations for the Hopfield model, the agreement between these two approaches can be tested—matching results are indicated in some cases. We also show that the present proposal exhibits a crucial attribute of living beings, which concerns their ability to react promptly to changes in the external environment. Additionally, we illustrate that this new approach may significantly enlarge the recognition capacity of neural networks in various situations; with correlated and non-correlated memories, as well as diluted, symmetric, or asymmetric interactions (synapses). This demonstrates that it can be implemented easily on a wide diversity of models.

Visual-motor contingency during symbol production contributes to short-term changes in the functional connectivity during symbol perception and long-term gains in symbol recognition

Letter production relies on a tight coupling between motor movements and visual feedback-each stroke of the letter is visually experienced as it is produced. Experience with letter production leads to increases in functional connectivity, a measure of neural communication, among visual and motor brain systems and leads to gains in letter recognition in preliterate children. We hypothesized that the contingency between the motor and visual experiences of the written form during production would result in both effects. Twenty literate adults were trained on four sets of novel symbols over the course of one week. Each symbol set was trained through one of four training conditions: drawing with ink, drawing without ink, watching a handwritten symbol unfold as if being drawn, and watching a static handwritten symbol. Contingency of motor and visual experiences occurred in the drawing with ink condition. The motor and visual experiences were rendered non-contingent in each of the other three conditions by controlling for visual or motor experience. Participants were presented with the trained symbols during fMRI scanning at three time points: one pre-training, one post-training, and one after a week-long no-training delay. Recognition was tested after each training session and after the third scan. We found that the contingency between visual and motor experiences during production changed the pattern of functional connectivity among visual, motor, and auditory neural communities and resulted in better recognition performance at post-training than at pre-training. Recognition gains were maintained after the no-training delay, but the functional connections observed immediately after training returned to their pre-training baselines. Our results suggest that behaviors that couple sensory and motor systems result in temporary changes in neural communication during perception that may not directly support changes in recognition.

From behavior to circuit modeling of light-seeking navigation in zebrafish larvae

Bridging brain-scale circuit dynamics and organism-scale behavior is a central challenge in neuroscience. It requires the concurrent development of minimal behavioral and neural circuit models that can quantitatively capture basic sensorimotor operations. Here, we focus on light-seeking navigation in zebrafish larvae. Using a virtual reality assay, we first characterize how motor and visual stimulation sequences govern the selection of discrete swim-bout events that subserve the fish navigation in the presence of a distant light source. These mechanisms are combined into a comprehensive Markov-chain model of navigation that quantitatively predicts the stationary distribution of the fish’s body orientation under any given illumination profile. We then map this behavioral description onto a neuronal model of the ARTR, a small neural circuit involved in the orientation-selection of swim bouts. We demonstrate that this visually-biased decision-making circuit can capture the statistics of both spontaneous and contrast-driven navigation.

All animals with the ability to move use sensory signals to help them navigate towards areas that seem better than their current location. Such areas might contain desirable things like food and mates, or they might allow an animal to escape from threats such as predators. But how the brain gives rise to this navigation behavior is unclear.

Karpenko et al. have now obtained insights into the underlying mechanism by studying a behavior in zebrafish larvae called phototaxis. Phototaxis is the tendency to move in response to light. The advantage of using zebrafish larvae to study this behavior is that their brains are small and semi-transparent. This makes it possible to record the activity of almost every neuron. As a result, an individual’s brain activity can be mapped on to their behavior more precisely than in most other species.

To probe how visual cues influence fish behavior, Karpenko et al. exposed individual fish to a carefully controlled virtual light source and then tracked their movements with a camera. The fish used two strategies to move towards the light. They selected their next movement based partly on the difference in the amount of light reaching each of their eyes, and partly on the change in overall brightness with each swim movement. Karpenko et al. used this information to build a numerical model of fish phototaxis, and to show how a simple brain circuit could generate this behavior.

Species whose brains differ in size and structure may nevertheless develop similar strategies to perform similar tasks. By quantifying a generic behavior in a simple animal model, this study could provide insights into comparable behaviors in other species. In addition, the study suggests a simple mechanism for how animals select actions on the basis of sensory signals, which may also be relevant to other species and other tasks.

Pattern reverberation in networks of excitable systems with connection delays

We consider the recurrent pulse-coupled networks of excitable elements with delayed connections, which are inspired by the biological neural networks. If the delays are tuned appropriately, the network can either stay in the steady resting state, or alternatively, exhibit a desired spiking pattern. It is shown that such a network can be used as a pattern-recognition system. More specifically, the application of the correct pattern as an external input to the network leads to a self-sustained reverberation of the encoded pattern. In terms of the coupling structure, the tolerance and the refractory time of the individual systems, we determine the conditions for the uniqueness of the sustained activity, i.e., for the functionality of the network as an unambiguous pattern detector. We point out the relation of the considered systems with cyclic polychronous groups and show how the assumed delay configurations may arise in a self-organized manner when a spike-time dependent plasticity of the connection delays is assumed. As excitable elements, we employ the simplistic coincidence detector models as well as the Hodgkin-Huxley neuron models. Moreover, the system is implemented experimentally on a Field-Programmable Gate Array.

Autocorrelation structure at rest predicts value correlates of single neurons during reward-guided choice

Correlates of value are routinely observed in the prefrontal cortex (PFC) during reward-guided decision making. In previous work (Hunt et al., 2015), we argued that PFC correlates of chosen value are a consequence of varying rates of a dynamical evidence accumulation process. Yet within PFC, there is substantial variability in chosen value correlates across individual neurons. Here we show that this variability is explained by neurons having different temporal receptive fields of integration, indexed by examining neuronal spike rate autocorrelation structure whilst at rest. We find that neurons with protracted resting temporal receptive fields exhibit stronger chosen value correlates during choice. Within orbitofrontal cortex, these neurons also sustain coding of chosen value from choice through the delivery of reward, providing a potential neural mechanism for maintaining predictions and updating stored values during learning. These findings reveal that within PFC, variability in temporal specialisation across neurons predicts involvement in specific decision-making computations.

DOI:http://dx.doi.org/10.7554/eLife.18937.001

Laminar differences in recurrent excitatory transmission in the rat entorhinal cortex in vitro

Paired intracellular recordings were used to investigate recurrent excitatory transmission in layers II, III and V of the rat entorhinal cortex in vitro. There was a relatively high probability of finding a recurrent connection between pairs of pyramidal neurons in both layer V (around 12%) and layer III (around 9%). In complete contrast, we have failed to find any recurrent synaptic connections between principal neurons in layer II, and this may be an important factor in the relative resistance of this layer in generating synchronized epileptiform activity. In general, recurrent excitatory postsynaptic potentials in layers III and V of the entorhinal cortex had similar properties to those recorded in other cortical areas, although the probabilities of connection are among the highest reported. Recurrent excitatory postsynaptic potentials recorded in layer V were smaller with faster rise times than those recorded in layer III. In both layers, the recurrent potentials were mediated by glutamate primarily acting at alpha-amino-3-hydroxy-5-methyl-4-isoxazole receptors, although there appeared to be a slow component mediated by N-methyl-D-aspartate receptors. In layer III, recurrent transmission failed on about 30% of presynaptic action potentials evoked at 0.2Hz. This failure rate increased markedly with increasing (2, 3Hz) frequency of activation. In layer V the failure rate at low frequency was less (19%), and although it increased at higher frequencies this effect was less pronounced than in layer III. Finally, in layer III, there was evidence for a relatively high probability of electrical coupling between pyramidal neurons. We have previously suggested that layers IV/V of the entorhinal cortex readily generate synchronized epileptiform discharges, whereas layer II is relatively resistant to seizure generation. The present demonstration that recurrent excitatory connections are widespread in layer V but not layer II could support this proposal. The relatively high degree of recurrent connections and electrical coupling between layer III cells may be a factor in it's susceptibility to neurodegeneration during chronic epileptic conditions.