Our ability to effectively retrieve complex semantic knowledge meaningfully impacts our daily lives, yet the interactions between semantic control and semantic representational systems that underly successful access and transient failures in access remain only partially understood. In this fMRI study, we contrast activation during successful semantic access, unsuccessful semantic access due to transient access-failures (i.e., ‘tip-of-the-tongue’, ‘feeling-of-knowing’), and trials where the semantic knowledge was not possessed. Twenty-four participants were presented 240 trivia-based questions relating to person, place, object or scholastic knowledge-domains. Whole brain analyses of the recall event indicate comparable recruitment of prefrontal semantic control systems during successful and unsuccessful semantic access and greater activation in representational systems in successful access. Region-of-interest analysis of domain-selective areas showed that successful access was generally associated with increased responses for both preferred and non-preferred stimuli, with the exception of place-selective regions (PPA, TOS and RSC). Both whole brain and Region-of-interest analysis showed the particular recruitment of place-selective regions during unsuccessful attempts at semantic access, for all stimulus domains. Collectively, these results suggest that prefrontal semantic control systems and classical spatial-knowledge-selective regions work together to locate relevant information and that access to complex knowledge results in a broadening of semantic representation to include regions selective for other knowledge domains.

Participants read sentences questioning their knowledge on a series of general-knowledge facts. Trials are divided into events where the knowledge was accessed, where access was unsuccessful yet participants felt they possessed the knowledge, and trials where the knowledge was not possessed.

This fMRI data has been slice-time corrected, realigned to correct for head movement, normalised to MNI space and smoothed using a 6mm FWHM isotropic kernel.

Domain-Knowledge Selective ROI and Brainmaps (DomainSelective_ROIs_BrainMaps.zip)

The ROIs folder contains filly nifti format voxel masks for each region ROI in table 2 of the manuscript (as detailed in the manuscript, only a subset of these ROIs was used in ROI analysis). The BrainMaps folder contains a functional nifti file of significant voxels for each domain-selective contrast

Full fMRI Data (WKAccess.zip)

Individual Subject Data

The data are organised by subject (1-24). Each folder contains the preprocessed data for each session and a 1st level GLM, motion parameters for each subject and the onsets, names and duration (‘multi condition’ file) for each subject for each trial.

SPM uses absolute paths so some analyses are not possible without updating the path information in the SPM.mat structure. Basic contrasts etc should be possible.

The helper scripts will allow these GLMs to be re-run, which will also update the path enabling full analyses.

Group Level Data ('Group_noMeta', 'Group_onlyMeta')

Two RFX GLMs are provided. One for the Recall-Event, the other for the Metacognitive Judgement (see manuscript for details).

Provided are the contrasts used to generate the figures in the manuscript.

As noted in the manuscript, contrasts examining the effect of access type contain extensive activation within regions associated with reward and these (and all) contrasts should be inclusively makes with CatSelp05.nii image-file, present in the ~/masks folder.

Helper Scripts

These scripts rely on relative paths and assumed path structure. The location/name of the 'root'  WKAccess folder is flexible but the scripts will not run if folders are renamed or files are moved relative to this.

StatMaster.m is a wrapper function to replicate the whole-brain analyses as reported in the manuscript. If this is run directly, it will replicate the whole-brain analysis in the manuscript (existing analysis folders must be renamed/moved/deleted first) .

It calls:

ORIG_GLM_batch_script_concatenate_2021.m

This performs the first level analysis concatenating the four sessions. It takes the two arguments: subject number; #scans per session

BatchContrastSingleSub.m

This performs 1st-level subject-specific contrasts that facilitate group-level analyses as well as some quality control contrasts.

This function takes four arguments: vector of subjects (i.e. [1:24]); name of the statistics folder where the SPM.mat is located, number of conditions to include; number of regressors not to include (e.g. motion parameters)

Group_analysis_script_noMeta.m and Group_analysis_script_onnlyMeta.m perform the second level analyses for Recall Event and the Metacognitive Judgement, respectively. Both functions take a vector of subject numbers (i.e. 1:24) as the input.