Performance of normative and approximate evidence accumulation on the dynamic clicks task

@article{Radillo2019PerformanceON,
  title={Performance of normative and approximate evidence accumulation on the dynamic clicks task},
  author={Adrian E. Radillo and Alan Veliz-Cuba and Kre{\vs}imir Josi{\'c} and Zachary P. Kilpatrick},
  journal={bioRxiv},
  year={2019}
}
The aim of a number of psychophysics tasks is to uncover how mammals make decisions in a world that is in flux. Here we examine the characteristics of ideal and near–ideal observers in a task of this type. We ask when and how performance depends on task parameters and design, and, in turn, what observer performance tells us about their decision-making process. In the dynamic clicks task subjects hear two streams (left and right) of Poisson clicks with different rates. Subjects are rewarded when… 
Human inference in changing environments
TLDR
It is argued that humans mimic Bayesian inference by approximating the posterior with a modest number of samples, and this results contribute to a growing literature on sample-based cognition and compression by stochastic pruning.
Analyzing dynamic decision-making models using Chapman-Kolmogorov equations
TLDR
It is shown that the probability density of the belief can be represented using differential Chapman-Kolmogorov equations, allowing efficient computation of ensemble statistics, and it is demonstrated that this method can be applied to tasks in which evidence arrives in a discrete, pulsatile fashion, rather than continuously.
Human inference in changing environments with temporal structure.
TLDR
This paper shows that humans adapt their inference process to fine aspects of the temporal structure in the statistics of stimuli, and identifies the nature of the deviations from Bayesian models that can account for the observed discrepancies from optimality in human behavior.
Normative Decision Rules in Changing Environments
TLDR
Adaptive decision strategies perform robustly even when implemented imperfectly (noisily) and can account for observed response times on a task with time-varying evidence better than commonly used constant-threshold or urgency-gating models.

References

SHOWING 1-10 OF 73 REFERENCES
Normative evidence accumulation in unpredictable environments
TLDR
A novel, normative formulation of adaptive learning models that forms decisions by acting as a leaky accumulator with non-absorbing bounds is presented, which provides new insights into the expectation-driven dynamics of the underlying neural signals.
Spatiotemporal dynamics of random stimuli account for trial-to-trial variability in perceptual decision making
TLDR
Using a new experimental visual tracking task and a novel Bayesian decision making model, it is found that the spatio-temporal noise fluctuations in the input of single trials explain a significant part of the observed responses.
The physics of optimal decision making: a formal analysis of models of performance in two-alternative forced-choice tasks.
TLDR
It is proved that there is always an optimal trade-off between speed and accuracy that maximizes various reward functions, including reward rate (percentage of correct responses per unit time), as well as several other objectives, including ones weighted for accuracy.
Rats adopt the optimal timescale for evidence integration in a dynamic environment
TLDR
The ability of rats to integrate and discount evidence provided by auditory clicks to infer a hidden, dynamic, state of the world and model the consequence of sensory noise to explain the source of errors are investigated.
Multialternative drift-diffusion model predicts the relationship between visual fixations and choice in value-based decisions
TLDR
A model of trinary value-based choice is proposed that generalizes what is known about binary choice, and is found to provide a quantitatively accurate description of the relationship between choice, reaction time, and visual fixation data using the same parameters that were estimated in previous work on binary choice.
Revisiting the Evidence for Collapsing Boundaries and Urgency Signals in Perceptual Decision-Making
TLDR
It is found that evidence for static or dynamic response boundaries may depend on specific paradigms or procedures, such as the extent of task practice, and the difficulty of selecting between collapsing and fixed bounds models has received insufficient attention in previous research.
Analyzing dynamic decision-making models using Chapman-Kolmogorov equations
TLDR
It is shown that the probability density of the belief can be represented using differential Chapman-Kolmogorov equations, allowing efficient computation of ensemble statistics, and it is demonstrated that this method can be applied to tasks in which evidence arrives in a discrete, pulsatile fashion, rather than continuously.
Rats and Humans Can Optimally Accumulate Evidence for Decision-Making
TLDR
This work developed decision-making tasks in which sensory evidence is delivered in randomly timed pulses, and analyzed the resulting data with models that use the richly detailed information of each trial’s pulse timing to distinguish between different decision- making mechanisms.
Visual Evidence Accumulation Guides Decision-Making in Unrestrained Mice
TLDR
By leveraging a large dataset to quantitatively characterize decision-making behavior, mice are established as suitable for neural circuit manipulation studies and it is demonstrated that mice accumulate visual evidence, demonstrating that this strategy for reducing uncertainty in decision- making is used by animals with diverse visual systems.
A bias–variance trade-off governs individual differences in on-line learning in an unpredictable environment
TLDR
It is shown that individual variability in learning from noisy evidence involves a bias–variance trade-off that is best explained by a model using a sampling algorithm that approximates optimal inference.
...
...