Neural Encoding by Bursts of Spikes

Access to files

•   FULL-TEXT.PDF (pdf)

Abstract

Neurons can respond to input by firing isolated action potentials or spikes. Sequences of spikes have been linked to the encoding of neuron input. However, many neurons also fire bursts; mechanistically distinct responses consisting of brief high-frequency spike firing. Bursts form separate response symbols but historically have not been thought to encode input. However, recent experimental evidence suggests that bursts can encode input in parallel with tonic spikes. The recognition of bursts as distinct encoding symbols raises important questions; these form the basic aims of this thesis: (1) What inputs do bursts encode? (2) Does burst structure provide extra information about different inputs. (3) Is burst coding robust against the presence of noise; an inherent property of all neural systems? (4) What mechanisms are responsible for burst input encoding? (5) How does burst coding manifest in in-vivo neurons. To answer these questions, bursting is studied using a combination of neuron models and in-vivo hippocampal neuron recordings. Models ranged from neuron-specific cell models to models belonging to three fundamentally different burst dynamic classes (unspecific to any neural region). These classes are defined using concepts from non-linear system theory. Together, analysing these model types with in-vivo recordings provides a specific and general analysis of burst encoding.For neuron-specific and unspecific models, a number of model types expressing different levels of biological realism are analysed. For the study of thalamic encoding, two models containing either a single simplified burst-generating current or multiple currents are used. For models simulating three burst dynamic classes, three further models of different biological complexity are used. The bursts generated by models and real neurons were analysed by assessing the input they encode using methods such as information theory, and reverse correlation. Modelled bursts were also analysed for their resilience to simulated neural noise.In all cases, inputs evoking bursts and tonic spikes were distinct. The structure of burst-evoking input depended on burst dynamic class rather than the biological complexity of models. Different n-spike bursts encoded different inputs that, if read by downstream cells, could discriminate complex input structure. In the thalamus, this n-spike burst code explains informative responses that were not due to tonic spikes. In-vivo hippocampal neurons and a pyramidal cell model both use the n-spike code to mark different LFP features. This n-spike burst may therefore be a general feature of bursting relevant to both model and in-vivo neurons.Bursts can also encode input corrupted by neural noise, often outperforming the encoding of single spikes. Both burst timing and internal structure are informative even when driven by strongly noise-corrupted input. Also, bursts induce input-dependent spike correlations that remain informative despite strong added noise. As a result, bursts endow their constituent spikes with extra information that would be lost if tonic spikes were considered the only informative responses.

Layman's Abstract

Neurons communicate with each other by generating brief electo-chemical action potentials or spikes. Typically when neurons receive an input, they respond by firing sequences of spikes with a firing rate reflecting this input; a process known as neural coding. However, neurons can also fire high-frequency groups of spikes called bursts. These bursts are distinct response events (generated by different cellular processes) and were traditionally thought not to play an important role in neural coding. However, recent experimental evidence is beginning to challenge this view and show that the mechanisms driving bursts may allow them to encode neural input.This thesis uses computer modelling and recordings taken from functioning neurons in-vivo to investigate the role bursts play in neural coding, with the following broad aims. (1) Find what neural input leads to the firing of bursts and analyse whether burst size is dependent on this input. (2) Neurons are inherently noisy, meaning their response will vary even when given identical inputs. Therefore, here a realistic study of burst encoding is undertaken by driving bursting models with input containing artificial noise. (3) Since a range of cells exhibit bursting behaviour, it is important to investigate whether burst coding varies between different cell types. To address this, the study of burst coding (aims 1 and 2) will be applied across a number of neural models ranging in complexity. A simple model consists of few basic equations set to mimic burst and spike responses - without accounting for any neural physiology, whereas a complex model may approximate the physiology of a real cell. In addition, (4) bursting models are separated by the most basic aspect of their behaviour - their dynamics. The final aim is to study how the encoding of stimuli by bursts and spikes is affected by the dynamics of the neuron.Modelled and in-vivo bursts of different sizes were dependent on neural input. Therefore, a neuron receiving these bursts could use them to decode the structure of the original neural input. Interestingly, the biological realism of the model was not important for coding. Indeed, all models studied showed burst coding. However, the input that evoked these bursts varied between models, specifically depending on the models dynamics. Therefore, burst coding is a direct result of neuron dynamics and can be accurately studied using simple models. It is thought that spike and burst activity depends on certain aspects of the preceding stimulation. For example, a burst may be thought to depend on a cell receiving a high amplitude stimulation. However, here it was found that bursts of different sizes were dependent on the overall structure of input stimuli rather than any one aspect of this stimulation (i.e. pre-burst amplitude). Therefore, bursts may be used to decode complex input. Burst encoding was also resistant to strong levels of neural noise, while single spikes were not. To summarise, contrary to previous understanding, bursts are important events in a neuron's response. They can encode input even in the presence of significant noise and their spike count can be used to denote different neural input, the structure of which can be predicted by neuron dynamics.

Additional content not available electronically

There is a digital repository of analysis scripts and neuron model programmes relevant to the thesis available at https://www.dropbox.com/sh/pjk67vz8w99okak/ETLTp-5oTu.

Keyword(s)

Burst coding; Bursts; Computational neuroscience; Hippocampus; Information theory; Local field potential; Neural coding; Neural dynamics; Neuron models; Neuron response; Thalamus

Bibliographic metadata

Institutional metadata

Record metadata