Skip to content

Project submission: Intersubject correlation analysis #4

@Nieto-CaballeroVE

Description

@Nieto-CaballeroVE

Group name: SynchroTales
Group members: Victor

Project description: Intersubject correlation (ISC) is a simple but powerful method used to identify shared neural signals between subjects. It effectively decomposes a signal into the sum of 3 components: a subject’s idiosyncratic signal, the shared signal with others, and an error term (Nastase et al., 2019). It works like this: you take the time series of one subject and correlate it with the average activity of all other subjects, per brain region. Once you do this for all subjects, you’ll have a correlation for each subject, which you can then average into one value. A high correlation value indicates that this region has a high shared component across subjects, and thus is driven by the stimulus. A critical assumption, however, is that the signals you collect for all subjects come from the same brain region. Otherwise, you would end up with more idiosyncratic and less shared signal components. Given that ECoG electrode coverage is different for each subject, this kind of analysis is more difficult. However, many subjects often have electrodes in similar places (e.g., superior temporal gyrus and inferior frontal gyrus). Thus, ISC can be performed by grouping electrodes based on location if there are enough electrodes amongst all subjects.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions