English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Decision-making is a cognitive process involving a choice between multiple alternatives. In recent years, there has been a lot of progress about understanding the way decisions are carried out in the brain. Neurons accumulate both evidence and noise over time, by a summation of spikes, in favor of a specific choice. It remains unclear, however, what are the noise sources involved in brain processing during cognitive tasks. Theoretical models often use an external noisy term added to the network to create trial-to-trial variability by adding random fluctuating inputs to the membrane potential of neurons [1]. Here, we propose an alternative method of information processing by investigating the influence of intrinsic noise generated through the complex interactions between cells in a model of spiking neurons performing a decision-making task. A paradigm termed the random dot motion task [2], used in many perceptual tasks, was incorporated in the artificial neural network to measure the model’s performance with two- or four-choice alternatives. In this task, dots presented on a visual display move in a specific direction and an added percentage of dots are moving randomly. The network must discriminate among the net movement of dots. Figure 1A-B illustrates the speed/accuracy trade-off of the modeled network using an internal noise source when presented with a dot motion of different motion strength values (varying degree of difficulty according to the ratio of moving dots). The model was exquisitely sensitive to small changes in initial conditions, and the addition of a single post-synaptic spike was able to bias the network’s decision in a systematic and predictable fashion. Figure ​Figure1C1C shows that a decision can be altered when the network is presented with a single additional spike in favor of the opposite choice, especially at low motion strength values. In sum, results of the model capture experimental findings on the electrophysiology of decision-making and suggest a key role of intrinsic noise on cognitive processing.    Figure 1  Speed & accuracy trade-off of a network using an internal noise source. A. Mean proportion of correctly selected choices relative to different motion strength values (0%, 3.2%, 6.4%, 12.8%, 25.6%, 51.2% & 100%) in a two (red) and four ...

Decision-making in a population of spiking neurons shaped by dynamics of intrinsic noise