# Wiki This repository contains the data and the analysis script used for the representational similarity analyses (RSA) in the publication **"Representational dynamics of memories for real-life events"** by Jeunehomme*, O., Heinen*, R., Stawarczyk, D., Axmacher, N. & D’Argembeau, A. (2022) [https://doi.org/10.1016/j.isci.2022.105391](https://doi.org/10.1016/j.isci.2022.105391) ### Abstract The continuous flow of experience that characterizes real-life events is not recorded as such in episodic memory but is condensed as a succession of event segments separated by temporal discontinuities. To unravel the neural basis of this representational structure, we recorded real-life events using wearable camera technology and used fMRI to investigate brain activity during their temporal unfolding in memory. We found that, compared to the representation of static scenes in memory, dynamically unfolding memory representations were associated with greater activation of the posterior medial episodic network. Strikingly, by analyzing the autocorrelation of brain activity patterns at successive time points throughout the retrieval period, we found that this network showed higher temporal dynamics when recalling events that included a higher density of event segments. These results reveal the key role of the posterior medial network in representing the dynamic unfolding of the event segments that constitute real-world memories. ## Files ### Behavioral analysis The script `Behavioral data_robust methods.R` used to analyze the behavioral data can be found in the **scripts** folder. For this you will also need to load `Rallfun-v39.txt` from [https://dornsife.usc.edu/labs/rwilcox/software/](https://dornsife.usc.edu/labs/rwilcox/software/) containing additional functions. Behavioral data `BehavioralData.csv` including gender, age, event densities, ect. are stored in **data/Behavioral_data** folder. ### Autobiographical brain mask The mask created using neurosynth, which was used for univariate and multivariate fMRI analyses, can be found in the **data** folder. ### Univariate analysis Contrast images for all univariate GLM contrast can be found in the **data/Univariate_GLM_contrasts** folder. ### Multivariate analysis - RSA Matlab code used to compute trialwise neural activity pattern similarity analyses `RepDyna_01_ROICorrs.m` and `RepDyna_02_ROIAnalyze.m` can be found in the **scripts** folder. Using `RepDyna_00_runAnalysis.m` we ran the functions used to generate the trialwise similarity values across all subjects. The script for the statistical analysis `RSA_Statistics.Rmd` and a knitted html version including all outputs `RSA_Statistics.html` can be found in the **scripts** folder. Processed data from the RSA analysis and the motion analysis can be found in the **data/ROI_and_motion_data** folder. This folder also contains the 6 motion parameters for each subject for the motion analysis. ### Multivariate analysis - Searchlight Using `RepDyna_00_runAnalysis.m` we ran the function used to generate the subject-wise similarity brain maps. Matlab code used to run the autocorrelation searchlight `RepDyna_03_searchlight.m` can be found in the **scripts** folder. This script generates autocorrelation similarity brain maps for each subject, that were then used for second-level group analysis using randomise (FSL) with tfce. Similarity brain maps are stored as `Autocorr_Sim_Maps.zip` in the **data/Searchlight_data** folder. Subject-wise similarity brain maps (z-spearman rho) can be found in the **data/Searchlight** folder. ### Raw data accessability Raw fMRI data can be accessed upon request by the authors. All other data and code needed to replicate the analyses can be made accessible by contacting the authors. ### Data files overview | analysis |file | content |type| |----------------|-------------------------------|-----------------------------|------| |behavior|`BehavioralData`| all behavioral data collected including subject gender and age|csv| |brain mask|`ROI_mask_autobiographical_memory_pFgA_z_FDR_0.01`|autobiographical network brain mask generated from neurosynth|nii| |univariate GLM|`actions_scenes`| actions vs control contrast image|nii| |univariate GLM|`actions_spatial`| actions vs spatial displacements contrast image|nii| |univariate GLM|`spatial_actions`| spatial displacements vs actions contrast image|nii| |univariate GLM|`spatial_scenes`| spatial displacements vs control contrast image|nii| |univariate GLM|`waiting_scenes`| waiting vs control contrast image|nii| |RSA|`01_OwnvsOtherTrials` |similarity values (z-spearman) reflecting the similarity of trials within conditions compared to trials of other conditions |csv| |RSA|`02_Recall2Cue_nodiagonal` |similarity values (z-spearman) reflecting the similarity of recall trials to the image cue of the same event compared to image cues of other events of the same condition (excluding the similarity to the image cue of the same trial) |csv| |RSA|`03_Repetition2Repetition`|similarity values (z-spearman) reflecting the similarity of each trial to the other three repetitions (same event, same condition)|csv| |motion|`04_motionParams`| 6 motion regressors from motion-correction preprocessing|csv| |searchlight|`AutocorrSimMaps`| brain maps for each subject and each condition containing the average autocorrelation between consecutive volumes during the recall phase. cond1=action, cond2=spatial, cond3=waiting, cond4=control|zip/nii.gz|