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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Studies investigating neural information processing often implicitly askboth, which processing strategy out of several alternatives is used and howthis strategy is implemented in neural dynamics. A prime example are studies onpredictive coding. These often ask if confirmed predictions about inputs orpredictions errors between internal predictions and inputs are passed on in ahierarchical neural system--while at the same time looking for the neuralcorrelates of coding for errors and predictions. If we do not know exactly whata neural system predicts at any given moment, this results in a circularanalysis--as has been criticized correctly. To circumvent such circularanalysis, we propose to express information processing strategies (such aspredictive coding) by local information-theoretic quantities, such that theycan be estimated directly from neural data. We demonstrate our approach byinvestigating two opposing accounts of predictive coding-like processingstrategies, where we quantify the building blocks of predictive coding, namelypredictability of inputs and transfer of information, by local activeinformation storage and local transfer entropy. We define testable hypotheseson the relationship of both quantities to identify which of the assumedstrategies was used. We demonstrate our approach on spiking data from theretinogeniculate synapse of the cat. Applying our local information dynamicsframework, we are able to show that the synapse codes for predictable ratherthan surprising input. To support our findings, we apply measures from partialinformation decomposition, which allow to differentiate if the transferredinformation is primarily bottom-up sensory input or information transferredconditionally on the current state of the synapse. Supporting our localinformation-theoretic results, we find that the synapse preferentiallytransfers bottom-up information.

Information-theoretic analyses of neural data to minimize the effect of researchers' assumptions in predictive coding studies