Version 1.0.0 This protocol has been created as support for the article: Yuki Kubota and Taiki Fukiage. (2025): Human-like monocular depth biases in deep neural networks, PLOS Computational Biology. **1. Overall** This file contains scripts created for three main purposes: DNN estimation outputs, participant experiment implementation, and data analysis. When using this code in academic papers, please cite our paper. **2. Environments / Requirements** Please check requirements for our code in sub-directories (models, experiments, analysis) **3. Directory Structures** Parts of our codes should be modified to appropriate path in reference to your directory structures. The codes in this directory only show how to analyze our data. - models - models-response-output.ipynb: Main estimation code. We recommend resetting the kernel and executing the predictions. - Models > [A] We recommend that you should download publicly available codes and pre-trained models as subdirectories. - val_data > nyu_test > image > [B] You should download NYU Depth v2 color images here. - experiments - css - csv - data - dist > [C] You should download and relocate jspsych libraries here. - image > nyu_test > image > [B] You should download NYU Depth v2 color images here. - index.html - instructions - js - analysis - analysis - main - abs_additional_analysis.ipynb: Additional analysis code (semantics, GT & human correlation) - abs_analysis.ipynb: Main analysis code - absRaw_analysis.ipynb: Main analysis code (before half-split) - abs_human_formatting.ipynb: Formatting human data - abs_model_formatting.ipynb: Formatting model data - graph_visualization.ipynb: Graph visualization - rank_analysis.ipynb: Rank analysis code - rankRaw_analysis.ipynb: Rank analysis code (before half-split) - sub - 2afc_analysis.ipynb: 2AFC analysis code - rel_analysis.ipynb: Relative data analysis code - rel_human_formatting.ipynb: Formatting supplemental human data - rel_model_formatting.ipynb: Formatting supplemental model data - csv - Humans-raw_abs_src.csv: Aggregated participants raw data - Models_abs_src.csv: Aggregated DNNs raw data - model_characteristics.csv: DNN characteristics (strategy, dataset etc.) - segmentation_label.csv: Class label and number correspondence table - val_image > [B] You should download NYU Depth v2 color images here. **4. How To Use the Codes** (a) model output (models) 1. Install the python libraries according to Requirements.txt. 2. Clone the repositories and download pre-trained DNNs to directory [A] (repository list is available at Supporting Information, S1 Table). 3. Download color images of NYU Depth V2 to directory [B] 4. Launch jupyter lab and execute the code for each model. (b) experiments (experiments) 1. Download color images of NYU Depth V2 to directory [B] 2. Launch a PHP website directly under the “experiments” repository. 3. Access http://localhost:8080/?c=1. You can start experiments under different conditions by specifying the “c” parameter from 1 to 46 – the URL at the beginning shows the case where c = 1. (c) data analysis (analysis) 1. Download color images of NYU Depth V2 to directory [B] 2. Launch jupyter lab and make sure the path in the file is valid. 3. Execute in the following order. (1) Formatting human and model data (abs_human_formatting.ipynb , abs_model_formatting.ipynb). Note that you don't need to create raw data. (2) Main analysis code (abs_analysis.ipynb), Rank analysis code (rank_analysis.ipynb), Raw data analysis (absRaw_analysis.ipynb, rankRaw_analysis.ipynb), Additional analysis code (abs_additional_analysis.ipynb), (3) Graph visualization (graph_visualization.ipynb) 4. For the supplemental data, please execute in the following order. (0) Formatting human and model data (rel_human_formatting.ipynb , rel_model_formatting.ipynb). Since aggregated raw data is stored, there is no need to execute this code. (1) Main analysis code (rel_analysis.ipynb), 2AFC analysis code (2afc_analysis.ipynb) (2) Graph visualization (graph_visualization.ipynb) **Author** All the code has been implemented by Yuki Kubota and Taiki Fukiage. **Copyright** © 2025 Yuki Kubota and Taiki Fukiage. This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). https://creativecommons.org/licenses/by-nc/4.0/