Dataset Information

Learning to make external sensory stimulus predictions using internal correlations in populations of neurons.

ABSTRACT: To compensate for sensory processing delays, the visual system must make predictions to ensure timely and appropriate behaviors. Recent work has found predictive information about the stimulus in neural populations early in vision processing, starting in the retina. However, to utilize this information, cells downstream must be able to read out the predictive information from the spiking activity of retinal ganglion cells. Here we investigate whether a downstream cell could learn efficient encoding of predictive information in its inputs from the correlations in the inputs themselves, in the absence of other instructive signals. We simulate learning driven by spiking activity recorded in salamander retina. We model a downstream cell as a binary neuron receiving a small group of weighted inputs and quantify the predictive information between activity in the binary neuron and future input. Input weights change according to spike timing-dependent learning rules during a training period. We characterize the readouts learned under spike timing-dependent synaptic update rules, finding that although the fixed points of learning dynamics are not associated with absolute optimal readouts they convey nearly all of the information conveyed by the optimal readout. Moreover, we find that learned perceptrons transmit position and velocity information of a moving-bar stimulus nearly as efficiently as optimal perceptrons. We conclude that predictive information is, in principle, readable from the perspective of downstream neurons in the absence of other inputs. This suggests an important role for feedforward prediction in sensory encoding.

SUBMITTER: Sederberg AJ

PROVIDER: S-EPMC5798322 | biostudies-literature | 2018 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Learning to make external sensory stimulus predictions using internal correlations in populations of neurons.

Sederberg Audrey J AJ MacLean Jason N JN Palmer Stephanie E SE

Proceedings of the National Academy of Sciences of the United States of America 20180118 5

To compensate for sensory processing delays, the visual system must make predictions to ensure timely and appropriate behaviors. Recent work has found predictive information about the stimulus in neural populations early in vision processing, starting in the retina. However, to utilize this information, cells downstream must be able to read out the predictive information from the spiking activity of retinal ganglion cells. Here we investigate whether a downstream cell could learn efficient encod ...[more]

PMID: 29348208

Similar Datasets

Project description:Stimulus-dependent expression of the retinoic acid-inactivating enzyme Cyp26B1 in olfactory sensory neurons (OSNs) forms a dorsomedial (DM)-ventrolateral (VL) gradient in the mouse olfactory epithelium. The gradient correlates spatially with different rates of OSN turnover, as well as the functional organization of the olfactory sensory map, into overlapping zones of OSNs that express different odorant receptors (ORs). Here, we analyze transgenic mice that, instead of a stimulus-dependent Cyp26B1 gradient, have constitutive Cyp26B1 levels in all OSNs. Starting postnatally, OSN differentiation is decreased and progenitor proliferation is increased. Initially, these effects are selective to the VL-most zone and correlate with reduced ATF5 expression and accumulation of OSNs that do not express ORs. Transcription factor ATF5 is known to stabilize OR gene choice via onset of the stimulus-transducing enzyme adenylyl cyclase type 3. During further postnatal development of Cyp26B1 mice, an anomalous DM(high)-VL(low) expression gradient of adenylyl cyclase type 3 appears, which coincides with altered OR frequencies and OR zones. All OR zones expand ventrolaterally except for the VL-most zone, which contracts. The expansion results in an increased zonal overlap that is also evident in the innervation pattern of OSN axon terminals in olfactory bulbs. These findings together identify a mechanism by which postnatal sensory-stimulated vitamin A metabolism modifies the generation of spatially specified neurons and their precise topographic connectivity. The distributed patterns of vitamin A-metabolizing enzymes in the nervous system suggest the possibility that the mechanism may also regulate neuroplasticity in circuits other than the olfactory sensory map.Significance statementThe mouse olfactory sensory map is functionally wired according to precise axonal projections of spatially organized classes of olfactory sensory neurons in the nose. The genetically controlled mechanisms that regulate the development of the olfactory sensory map are beginning to be elucidated. Little is known about mechanisms by which sensory stimuli shape the organization of the map after birth. We show that a stimulus-dependent gradient of a retinoic acid-inactivating enzyme Cyp26B1 modifies the composition, localization, and axonal projections of olfactory sensory neuron classes. The mechanism is novel and suggests the interesting possibility that local vitamin A metabolism could also be a mediator of stimulus-dependent modifications of precise spatial connectivity in other parts of the nervous system.

Dataset Information

Learning to make external sensory stimulus predictions using internal correlations in populations of neurons.

Publications

Learning to make external sensory stimulus predictions using internal correlations in populations of neurons.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets