<h1>Object Identifiers</h1> <h2>Task IDs</h2> <h3>Outside scanner</h3> - **T01:** Scene-Location matching task performed prior to scanning;<br /> - **T05:** Scene-Location matching task performed after scanning; - **T06:** Scene-Video matching task performed after scanning; <h3>Inside scanner</h3> - **LXXx:** Scene vs Face localiser task (oddball detection); - **T02a:** First run of in-scanner endpoint presentations prior to seeing the videos (oddball detection); - **T02b:** Second run of in-scanner endpoint presentations prior to seeing the videos (oddball detection); - **T03x:** Video task (discriminate overlap vs no-overlap videos); - **T04a:** First run of in-scanner endpoint presentations after seeing the videos (oddball detection); - **T04b:** Second run of in-scanner endpoint presentations after seeing the videos (oddball detection); <br/><br/> <h2>Stimuli IDs</h2> <h3>Factions</h3> Sets of endpoints containing 3 pairs of images from different locations are referred to as Factions and labelled as <b>[F1,F2,F3,F4]</b>. <h3>Endpoint images</h3> Endpoint images themselves are labelled in two different ways...<br /> <b>[A1,A2,B1,B2,C1,C2]</b> are identifiers that account for the image counterbalancing across conditions such that: - [A1,A2] are always in the overlap condition; - [B1,C2] are always in the no-overlap condition; - [C1,B2] are always in the unseen condition; <b>[X1,X2,Y1,Y2,Z1,Z2]</b> are identifiers that do not account for the image counterbalancing across conditions but refer to specific images per se. For all endpoint identifiers, the numeral denotes the left or right hand side of the panorama (1 and 2 respectively). <h3>Linking structures</h3> Linking structures <b>[L1,L2,L3]</b> are counterbalancing schemes that determine how endpoints are assigned to the experimental conditions:<br /> - For <b>L1</b>, [X1,X2] are assigned to the overlap condition, [Z1,Y2] are assigned to the no-overlap condition, and [Y1,Z2] are assigned to the unseen condition; - For <b>L2</b>, [Y1,Y2] are assigned to the overlap condition, [X1,Z2] are assigned to the no-overlap condition, and [Z1,X2] are assigned to the unseen condition; - For <b>L3</b>, [Z1,Z2] are assigned to the overlap condition, [Y1,X2] are assigned to the no-overlap condition, and [X1,Y2] are assigned to the unseen condition; Each participant was assigned one linking structure across all factions. <h3>Panning direction</h3> The direction of the camera pans in each panoramic video is denoted as follows: - <b>D1</b> (left to right); - <b>D2</b> (right to left); These directions are counterbalanced across factions and participants. <br/><br/> <h2>ROI IDs</h2> - <b>rPHC</b>: Right parahippocampal cortex; - <b>lPHC</b>: Left parahippocampal cortex; - <b>rRSC</b>: Right retrosplenial cortex; - <b>lRSC</b>: Left retrosplenial cortex; All ROIs are based off a first-level model of LXXx designated <b>N0_00B</b> (N0 denotes native space). <br/><br/> <h2>Analysis IDs</h2> <h3>Behavioural analyses</h3> - <b>Glme01</b>: Model of Scene-Location matching performance in T01 and T05 as a function of session (pre- vs post-videos), experimental condition (overlap, no-overlap vs unseen); - <b>Glme02</b>: Model of responses to B1 and C2 cues in T01 and T05 to test whether participants were matching B1 and C2 endpoints during these tasks; - <b>Glme03</b>: Model of Scene-Video matching performance in T06 as a function of condition (overlap vs no-overlap); <h3>ROI based analyses</h3> - <b>X04A</b>: Tests for representations of specific endpoints that remained constant across scanning sessions (i.e. image identity). Built from first-level models of T02* and T04* designated Alpha01. - <b>X05R</b>: Tests for representations of specific endpoints that remained constant within (but not necessarily across) scanning sessions (i.e. image identity). Built from first-level models of T02* and T04* designated Alpha01. - <b>X01R</b>: Test for view-independent representations of specific locations (i.e. location-based) within each a priori ROI. Built from first-level models of T02* and T04* designated Alpha01. - <b>Lmm_UniBoldEndpoints</b>: Tests for differences in univariate BOLD during T02* and T04* as function of session (pre- vs post-videos), experimental condition (overlap, no-overlap vs unseen), and behavioural performance in T05. Built from first-level models of T02* and T04* designated Alpha01. - <b>Lmm_UniBoldVideos</b>: Tests for differences in univariate BOLD during T03x as a function of video type (overlap vs no-overlap) and behavioural performance in T05. Built from first-level models of T03x designated <b>N0_01A</b>. <h3>Searchlight based RSA</h3> - <b>C00</b>: Test for representations of specific images (i.e. image identity) across the whole brain using spherical searchlights with radius = 3 voxels. Built from first-level models of T02* and T04* designated Alpha01. - <b>C01</b>: Test for view-independent representations of specific locations (i.e. location-based) across the whole brain using spherical searchlights with radius = 3 voxels. Built from first-level models of T02* and T04* designated Alpha01. <h3>Mass univariate analyses</h3> - <b>G1_K8_02A/2x3_BCov</b>: Tests for differences in univariate BOLD during T02* and T04* as function of session (pre- vs post-videos), experimental condition (overlap, no-overlap vs unseen), and behavioural performance in T05. Built from first-level models of T02* and T04* designated <b>02A</b>. - <b>G1_K8_01A/Overlap-Nooverlap_BCov</b>: Tests for differences in univariate BOLD during T03x as a function of video type (overlap vs no-overlap) and behavioural performance in T05. Built from first-level model of overlap vs no-over videos designated <b>01A</b>. Across these identifiers, the <b>G1</b> tag denotes that EPI data were normalised to MNI space using a DARTEL group template designated G1. <b>K8</b> denotes that an 8mm isotropic Gaussian smoothing kernel has been applied to the EPI data. <br /><br /><br /><br /> <h1>Analysis overview</h1> The 'Analyses+Results' folder contains 8 subdirectories: <br /> - <b>#_SimilarityData</b><br /> - <b>0_Misc</b><br /> - <b>1_Behaviour</b><br /> - <b>2_RSA_ViewpointSpecific</b><br /> - <b>3_RSA_LocationBased</b><br /> - <b>4_rPHC-vs-rRSC</b><br /> - <b>5_UniBold_Endpoints</b><br /> - <b>6_UniBold_Videos</b><br /> <h3>Similarity data and miscellaneous</h3> <b>#_SimilarityData</b> contains spreadsheets of the raw similarity data (Fisher-z scores) for all participants and ROIs. These spreadsheets also include plots of the mean similarity scores and confidence intervals estimated from the raw data.<br /> <br /> <b>0_Misc</b> contains miscellaneous items relevant to this project. It includes a representational similarity analysis (specifically X01R) of BOLD patterns in the occipital place area (OPA). This region was not the focus of a priori hypotheses and so is only briefly discussed in the manuscript. However, a supplementary analysis of OPA representations is provided here for reference. The 0_Misc directory also includes a dedicated analysis (labeled as X09R) testing for changes in baseline levels of representational similarity across sessions (i.e. T02 vs T04). This session effect is tested in the Genu of the Corpus callosum, a white matter region where BOLD effects are assumed to be minimal. 0_Misc additionally includes group-normalised masks of each a priori ROI for illustrative purposes. Finally, 0_Misc includes some MATLAB function that are dependencies of other scripts available in this project. Basically, think of the 0_Misc directory as the man-drawer of this project.<br /> <h3>Behaviour</h3> <b>1_Behaviour</b> contains all the behavioural analyses reported in the study. A subdirectory named <b>InScanner</b> lists discrimination accuracy (d') analyses relating to each of the in scanner tasks (LXXx, T02a, T02b, T03x, T04a, and T04b). 1_Behaviour also contains a subdirectory named <b>OutScanner</b> that includes analyses Glme01, Glme02, and Glme03. <h3>Viewpoint specific RSA</h3> <b>2_RSA_ViewpointSpecific</b> contains the analyses X04A, X05R and C00. These test for representations of specific images (i.e., image identity). <h3>Location-based RSA</h3> <b>3_RSA_LocationBased</b> contains the analyses X01R and C01. These test for view-independent representations of specific locations (i.e., location-based). <h3>rPHC versus rRSC</h3> <b>4_rPHC-vs-rRSC</b> details analyses that test for functional dissociations between rPHC and rRSC. This includes contrasts between different model terms in X01R, as well as instances of X01R run on subsets of the similarity data. <h3>Univariate analysis of endpoints images</h3> <b>5_UniBold_Endpoints</b> contains analyses Lmm_UniBoldEndpoints and G1_K8_02A/2x3_BCov. These test for differences in univariate BOLD during T02* and T04* as function of session (pre- vs post-videos), experimental condition (overlap, no-overlap vs unseen), and behavioural performance in T05. <h3>Univariate analysis of videos</h3> <b>5_UniBold_Videos</b> contains analyses Lmm_UniBoldVideos and G1_K8_01A/Overlap-Nooverlap_BCov. These test for differences in univariate BOLD during T03x as a function of video type (overlap vs no-overlap) and behavioural performance in T05.