Search for: All records

Accurately decoding external variables from observations of neural activity is a major challenge in systems neuroscience. Bayesian decoders, that provide probabilistic estimates, are some of the most widely used. Here we show how, in many common settings, the probabilistic predictions made by traditional Bayesian decoders are overconfident. That is, the estimates for the decoded stimulus or movement variables are more certain than they should be. We then show how Bayesian decoding with latent variables, taking account of low-dimensional shared variability in the observations, can improve calibration, although additional correction for overconfidence is still needed. Using data from males, we examine: 1) decoding the direction of grating stimuli from spike recordings in primary visual cortex in monkeys, 2) decoding movement direction from recordings in primary motor cortex in monkeys, 3) decoding natural images from multi-region recordings in mice, and 4) decoding position from hippocampal recordings in rats. For each setting we characterize the overconfidence, and we describe a possible method to correct miscalibration post-hoc. Properly calibrated Bayesian decoders may alter theoretical results on probabilistic population coding and lead to brain machine interfaces that more accurately reflect confidence levels when identifying external variables. Significance Statement Bayesian decoding is a statistical technique for making probabilistic predictions about external stimuli or movements based on recordings of neural activity. These predictions may be useful for robust brain machine interfaces or for understanding perceptual or behavioral confidence. However, the probabilities produced by these models do not always match the observed outcomes. Just as a weather forecast predicting a 50% chance of rain may not accurately correspond to an outcome of rain 50% of the time, Bayesian decoders of neural activity can be miscalibrated as well. Here we identify and measure miscalibration of Bayesian decoders for neural spiking activity in a range of experimental settings. We compare multiple statistical models and demonstrate how overconfidence can be corrected. 

Abstract In many areas of the brain, neural spiking activity covaries with features of the external world, such as sensory stimuli or an animal's movement. Experimental findings suggest that the variability of neural activity changes over time and may provide information about the external world beyond the information provided by the average neural activity. To flexibly track time-varying neural response properties, we developed a dynamic model with Conway-Maxwell Poisson (CMP) observations. The CMP distribution can flexibly describe firing patterns that are both under- and overdispersed relative to the Poisson distribution. Here we track parameters of the CMP distribution as they vary over time. Using simulations, we show that a normal approximation can accurately track dynamics in state vectors for both the centering and shape parameters (λ and ν). We then fit our model to neural data from neurons in primary visual cortex, “place cells” in the hippocampus, and a speed-tuned neuron in the anterior pretectal nucleus. We find that this method outperforms previous dynamic models based on the Poisson distribution. The dynamic CMP model provides a flexible framework for tracking time-varying non-Poisson count data and may also have applications beyond neuroscience. 

Many controlledin vitrostudies have demonstrated how postsynaptic responses to presynaptic spikes are not constant but depend on short-term synaptic plasticity (STP) and the detailed timing of presynaptic spikes. However, the effects of short-term plasticity (depression and facilitation) are not limited to short, subsecond timescales. The effects of STP appear on long timescales as changes in presynaptic firing rates lead to changes in steady-state synaptic transmission. Here, we examine the relationship between natural variations in the presynaptic firing rates and spike transmissionin vivo. Using large-scale spike recordings in awake male and female mice from the Allen Institute Neuropixels dataset, we first detect putative excitatory synaptic connections based on cross-correlations between the spike trains of millions of pairs of neurons. For the subset of pairs where a transient, excitatory effect was detected, we use a model-based approach to track fluctuations in synaptic efficacy and find that efficacy varies substantially on slow (∼1 min) timescales over the course of these recordings. For many connections, the efficacy fluctuations are correlated with fluctuations in the presynaptic firing rate. To understand the potential mechanisms underlying this relationship, we then model the detailed probability of postsynaptic spiking on a millisecond timescale, including both slow changes in postsynaptic excitability and monosynaptic inputs with short-term plasticity. The detailed model reproduces the slow efficacy fluctuations observed with many putative excitatory connections, suggesting that these fluctuations can be both directly predicted based on the time-varying presynaptic firing rate and, at least partly, explained by the cumulative effects of STP. SIGNIFICANCE STATEMENTThe firing rates of individual neurons naturally vary because of stimuli, movement, and brain state. Models of synaptic transmission predict that these variations in firing rates should be accompanied by slow fluctuations in synaptic strength because of short-term depression and facilitation. Here, we characterize the magnitude and predictability of fluctuations in synaptic strengthin vivousing large-scale spike recordings. For putative excitatory connections from a wide range of brain areas, we find that typical synaptic efficacy varies as much as ∼70%, and in many cases the fluctuations are well described by models of short-term synaptic plasticity. These results highlight the dynamic nature ofin vivosynaptic transmission and the interplay between synaptic strength and firing rates in awake animals. 

Synapses change on multiple timescales, ranging from milliseconds to minutes, due to a combination of both short- and long-term plasticity. Here we develop an extension of the common generalized linear model to infer both short- and long-term changes in the coupling between a pre- and postsynaptic neuron based on observed spiking activity. We model short-term synaptic plasticity using additive effects that depend on the presynaptic spike timing, and we model long-term changes in both synaptic weight and baseline firing rate using point process adaptive smoothing. Using simulations, we first show that this model can accurately recover time-varying synaptic weights (1) for both depressing and facilitating synapses, (2) with a variety of long-term changes (including realistic changes, such as due to STDP), (3) with a range of pre and postsynaptic firing rates, and (4) for both excitatory and inhibitory synapses. We then apply our model to two experimentally recorded putative synaptic connections. We find that simultaneously tracking fast changes in synaptic weights, slow changes in synaptic weights, and unexplained variations in baseline firing is essential. Omitting any one of these factors can lead to spurious inferences for the others. Altogether, this model provides a flexible framework for tracking short- and long-term variation in spike transmission.