A Reduced-Dimension fMRI Shared Response Model
Authors: Po-Hsuan (Cameron) Chen, Janice Chen, Yaara Yeshurun, Uri Hasson, James Haxby, Peter J. Ramadge
NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | 4 Experiments We assess the performance and robustness of SRM using f MRI datasets (Table 1) collected using different MRI machines, subjects, and preprocessing pipelines. |
| Researcher Affiliation | Academia | 1Department of Electrical Engineering, Princeton University 2Princeton Neuroscience Institute and Department of Psychology, Princeton University 3Department of Psychological and Brain Sciences and Center for Cognitive Neuroscience, Dartmouth College |
| Pseudocode | No | The paper describes the model and estimation process using mathematical equations and descriptions, but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access to source code (e.g., specific repository link, explicit code release statement, or code in supplementary materials) for the methodology described. |
| Open Datasets | Yes | f MRI datasets (Table 1) collected using different MRI machines, subjects, and preprocessing pipelines. The sherlock dataset was collected while subjects watched an episode of the BBC TV series Sherlock (66 mins).The raider dataset was collected while subjects viewed the movie Raiders of the Lost Ark (110 mins)... |
| Dataset Splits | Yes | We record the average accuracy and standard error by two-fold cross-validation over the data halves and leave-one-out over subjects. |
| Hardware Specification | No | The paper mentions 'f MRI datasets collected using different MRI machines' for data acquisition, but does not provide specific details (model numbers, processor types, memory amounts) about the hardware used to run the computational experiments. |
| Software Dependencies | No | The paper mentions the 'Fast ICA implementation [20]' but does not provide specific version numbers for any software libraries, frameworks, or solvers used in their experiments. |
| Experiment Setup | Yes | Experiment 1: SRM and spatial smoothing. We first use spatial smoothing to determine if we can detect a shared response in PMC for the sherlock dataset. ... We repeat this for five random subject divisions and average the results. ... Gaussian filter, with width at half height of 3, 4, 5 and 6mm... With k = 813 there is no reduction of dimension and SRM achieves a correlation equivalent to 6mm spatial smoothing. ... Learning 50 features achieves a 33% higher average correlation... Experiment 3: ... a linear SVM classifier trained on labeled voxel space data... When using k1 = 10 and k2 = 100, we observe significant improvement... |