Neuroimage April 2020 PMID 31866165

Inverse transformed encoding models - a solution to the problem of correlated trial-by-trial parameter estimates in fMRI decoding.

Joram Soch, Carsten Allefeld, John-Dylan Haynes

Download social card

Mentions

Peak: 24 in Jan 2020

across 2 sources since Apr 2019

Extended reach

236,390

potential audience

Total likes

Video views

105

Mentions per month

Apr 2019 – Nov 2023

Where it's discussed

X / Twitter35 posts
YouTube1 videos

AI synopsis

What people are saying

Several posts highlight the novelty and practical value of the inverse transformed encoding models, describing it as 'a solution to the problem of correlated trial-by-trial estimates.' The choice of words like 'solution,' 'final journal PDF,' and 'open access' emphasize both their interest and belief in its potential impact. The tone is generally professional and enthusiastic, with some posts actively encouraging further exploration or dissemination of the research, reflecting deep engagement and recognition of its importance.

AgreementModerate agreement

Most posts acknowledge the significance of the publication, with some expressing support or recognizing it as a notable advancement in fMRI decoding research.

InterestHigh interest

The discussions show a high level of curiosity, especially with statements like 'our solution' and mentions of open access, reflecting genuine enthusiasm and relevance to ongoing research.

EngagementHigh engagement

Numerous posts reference the methodology (inverse transformed encoding models) and its implications, indicating active engagement through sharing links, summaries, and personal comments.

ImpactModerate impact

While the posts suggest potential for meaningful impact, they predominantly focus on the technical advancement and its availability, implying a perceived but not yet fully realized high impact.

Abstract in brief

Inverse Transformed Encoding Models (ITEM) address the problem of highly variable and correlated trial-by-trial parameter estimates in fast event-related fMRI experiments, especially when trials are close together in time.
ITEM works by properly accounting for serial dependencies introduced by the hemodynamic response function, leading to more accurate decoding of experimental conditions or variables from brain activity without increasing computational load.
Through simulations and real data analysis, the authors show that ITEM outperforms existing methods in decoding accuracy and can even reconstruct visual information from fMRI signals effectively.]

Read the full abstract

All mentions

Inverse Transformed Encoding Models Improve fMRI Trial-Wise Decoding Precision

Inverse Transformed Encoding Models address trial-by-trial parameter estimate correlations in fast event-related fMRI, improving decoding accuracy and reconstructing visual information effectively.

Joram Soch · Oct 25, 2023 105 views

Joram Soch

@JoramSoch · X

In this #PaperVideo, I'm summarizing the 2020 NeuroImage paper about our solution for the problem of correlated trial-by-trial estimates in fMRI decoding. Don't suppress the correlations - estimate their extent and explain them away! #ITEM #fMRI https://t.co/A19e3Pd5s2

Nov 6, 2023 · 486 followers View post →

City Research Online

@City_Research · X

#openaccess New research in CRO: July 12, 2021 at 02:08PM Inverse Transformed Encoding Models – a solution to the problem of correlated trial-by-trial parameter estimates in fMRI decoding https://t.co/h2KGFjroa3

Jul 12, 2021 · 1,422 followers View post →