Deutsch
 
Benutzerhandbuch Datenschutzhinweis Impressum Kontakt
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT

Freigegeben

Poster

Hippocampal neural events predict ongoing brain-wide BOLD activity

MPG-Autoren
/persons/resource/persons75278

Besserve,  M
Max Planck Institute for Biological Cybernetics, Max Planck Society;
Department Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Dept. Empirical Inference, Max Planck Institute for Intelligent Systems, Max Planck Society;

/persons/resource/persons84063

Logothetis,  NK
Max Planck Institute for Biological Cybernetics, Max Planck Society;
Department Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Externe Ressourcen

Link
(beliebiger Volltext)

Volltexte (frei zugänglich)
Es sind keine frei zugänglichen Volltexte verfügbar
Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar
Zitation

Besserve, M., & Logothetis, N. (2016). Hippocampal neural events predict ongoing brain-wide BOLD activity. Poster presented at 46th Annual Meeting of the Society for Neuroscience (Neuroscience 2016), San Diego, CA, USA.


Zitierlink: http://hdl.handle.net/21.11116/0000-0000-7AD0-C
Zusammenfassung
Transient Local Field Potential (LFP) activity exhibits a wide variety of patterns, reflecting local antagonistic or synergistic neural activity changes in the recorded structure. Among them, “events” with a characteristic time-frequency profile can be identified, which may occasionally reflect transient large-scale interactions with other brain structures. Such an event-related multistructure activity can be studied using concurrent fMRI and LFP recordings in an experimental design dubbed as Neural Event Triggered (NET)-fMRI (Logothetis et al, 2012). Recently we used NET-fMRI to describe the brainwide up and down modulation of neural activity associatiated with hippocampal ripples. Here we examine how much the BOLD changes associated with these events can describe the fMRI time series along the entire data acquisition period. To address this question, we develop a generative model of the ongoing neural activity including both hippocampal LFP recordings and BOLD signals in the whole brain. The model was based on 6 types of oscillations detected in the hippocampus: Sharp-waves, ripples, gamma, beta, sigma and low frequency events. We first estimated the time course of the BOLD signature of neural events across brain structures by learning a dictionary of responses using the kSVD algorithm, and performed statistical analysis to extract significantly activated voxels for each neural event. Based on a convolutive model of the LFP-BOLD relationship, we corrected the effects of overlap between successive neural events on the BOLD response by estimating the autocorrelation function of the neural events and used it to obtain deconvolved BOLD signatures, describing the contribution of a single event to the BOLD signal. Preliminary results on 3 sessions show the BOLD signature of each event can be well captured by two dictionaries elements: one with a short response latency (peak response at 2.6s) in a wide range of subcortical and cortical structures, and a long latency (peak response at 5.1s) response restricted to sensory and associative cortical areas. This model enables us to estimate the contribution of hippocampus-related activity to fMRI time series in the whole brain. We thus estimated the overall ongoing single trial fMRI activity averaged across all brain structures at each time point using the model and LFP event time stamps only. The average correlation coefficient between the true fMRI signal and the event based reconstruction was .313, showing that hippocampal neural event carry rich information about global brain dynamics and suggesting that global brain dynamics could in turn be used to infer electrical activity non-invasively.