Behavioural data, A Task-Invariant Prior Explains Trial-by-Trial Active Avoidance Behaviour Across Gain and Loss Tasks.

# Bayesian priors in active avoidance Author: Tobias Granwald (orcid: 0009-0002-0208-6891) The folder contains the extracted data used in the analysis with the manuscript "A Task-Invariant Prior Explains Trial-by-Trial Active Avoidance Behaviour Across Gain and Loss Tasks." (Granwald et al., 2025). Data was collected and analysed in accordance with our preregistration: https://osf.io/rej74 The extracted data is the raw choice data from the robber task and the factory task in with all participants choices in one file. The following changes has been made in this file compared to the raw data file beyond putting all the data in to a single file: - Participant IDs have been anonymized by removing participant's prolific IDs and instead adding a numbers between 1 and 279. - Participant's responses in the PANAS and and STAI have been extracted and added as a separate row to the data. - Demographic data extracted from Prolific has been added as separate rows in the data. - Default data registered by jsPsych (de Leeuw, 2015) that was not of relevance to the tasks has been removed (trial_type, trial_index, time_elapsed, stimulus, response, timeout, failed_images, failed_audio, failed_video, question_order, correct) - Data about other unrelated tasks was removed (oMST; Stark et al., 2023). - Data from practice trials were removed. - Data from instruction trials or non-choice parts of the trials were removed. extractedData.csv contains the following collumns of data: - subjID_anon (integer, anonomized participant IDs, 1 to 279) - rt (integer; response time in ms) - response (string, keyboard response made by the participants "arrowleft" and "arrowright") - category (string, category data in the row, here selected only category "choice") - success (integer, binary indicating if the active action was successful, success = 1, loss or passive action = 0, participant failed to make a response = "noResponse") - robberTotPoints (integer, cummulative points earned during the current block of the robber task, trials of the factory task = 0 and start each block at = 0) - factoryTotPoints (integer, cummulative points earned during the current block of the factory task, trials of the robber task = 0 and start each block at = 0) - trial (integer, trial number of the tasks separtatly 0 to 59 for the first block and 60 to 119 for the second block) - task (string, in which task the current trial is played, robber task = "robber" and factory task = "factory") - offers (integer, offers presented to the participants in the tasks. In the robber task, the offer was the amount of the tip that the participants may earn at the end of the night before encountering the robber. In the factory task, the offers was the possible loss that the participant incurs if they let the machine break.) - cost (integer, the cost of the active action on the current trial) - noActValue (integer, the value of the passive action, in the robber task = 0, in the factory task = -offers) - stimNo (string, file name of the stimulus presented on the current trial robber task ex. "img/robber1.png" and factory task ex. "img/factory1.png") - probWin (floating-point, success probability used on the current trial to calculate based on a binomial distribution whether the participant would be successful or not in avoiding the negative outcome when chosing the active action) - act (integer, binary indicating whether the participant chose the active = 1 or the passive action = 0 or failed to make a response = "noResponse") - PosAffect_tot (integer, sum score for the participant's score in the positive affect subscale of the PANAS (Watson et al., 1988)) - STAIT_tot (integer, sum score for the participant's score in the trait subscale of the STAI Y-form (Spielberger et al., 1983)) - demograph_age (integer, the "Age" variable extracted from the demographic data provided by prolific.com age in years) - demograph_sex (string, the "Sex" variable extracted from the demographic data provided by prolific.com, male = "Male" and female = "Female") - demograph_nationality (string, "Nationality" variable extracted from the demographic data provided by prolific.com) - demograph_approvals (integer, "Total.approvals" variable extracted from the demographic data provided by prolific.com) - failed_AC_Q (integer, number of failed attention checks the participant had in the questionnaires (0 to 6), if more than 2 participants were excluded) extractedData_FU.csv contain the same collumns as the data above except for "demograph_approvals" and it instead includes: - demograph_dateDif (integer, the number of days between the first session and the follow-up session defined by "Completed.at" - "Completed.at" for the two sessions extracted from the demographic data provided by prolific.com) The task was programmed by Tobias Granwald at Karolinska Institutet utilizing JavaScript, html and css to ensure that data can be collected online, it also uses the toolbox: jsPsych v 7.1.0 (de Leeuw, 2015). The task can be found at https://github.com/Granwald/healthyPriors ## References de Leeuw, J. R. (2015). jsPsych: A JavaScript library for creating behavioral experiments in a Web browser. Behavior Research Methods, 47(1), 1–12. https://doi.org/10.3758/s13428-014-0458-y Granwald, T., Dayan, P., Lengyel, M., & Guitart-Masip, M. (2025). A Task-Invariant Prior Explains Trial-by-Trial Active Avoidance Behaviour Across Gain and Loss Tasks. Communications Psychology, 82(3) 1-14. https://doi.org/10.1038/s44271-025-00254-1 Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). State-Trait Anxiety Inventory for Adults Sampler Set Manual, Instrument and Scoring Guide. www.mindgarden.com. Stark, C. E. L., Noche, J. A., Ebersberger, J. R., Mayer, L., & Stark, S. M. (2023). Optimizing the mnemonic similarity task for efficient, widespread use. Frontiers in Behavioral Neuroscience, 17. https://www.frontiersin.org/articles/10.3389/fnbeh.2023.1080366 Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and Validation of Brief Measures of Positive and Negative Affect: The PANAS Scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063# 主动回避行为中的贝叶斯先验 作者：托比亚斯·格兰瓦尔德（Tobias Granwald，ORCID：0009-0002-0208-6891） 本文件夹包含用于分析的提取后数据，对应论文《一项任务不变先验解释跨收益与损失任务的逐试次主动回避行为》（Granwald等人，2025）。 本研究的数据收集与分析严格遵循预注册方案：https://osf.io/rej74 本次提取的数据为抢劫任务（robber task）与工厂任务（factory task）的原始选择数据，所有被试的试次选择均整合至单个文件内。 以下是本文件相较于原始数据所做的调整（除将所有数据整合至单个文件外）： - 被试ID已完成匿名化处理：移除原有的Prolific ID，替换为1至279的数字编号。 - 提取被试在积极与消极情感量表（PANAS, Positive and Negative Affect Schedule）与状态-特质焦虑量表（STAI, State-Trait Anxiety Inventory）中的作答结果，作为单独行添加至数据中。 - 提取自Prolific平台的人口统计学数据，以单独行的形式添加至数据集中。 - 删除与任务无关的、由jsPsych（de Leeuw, 2015）记录的默认数据字段，包括：trial_type、trial_index、time_elapsed、stimulus、response、timeout、failed_images、failed_audio、failed_video、question_order、correct。 - 删除与其他无关任务相关的数据（如记忆相似性任务oMST；Stark等人，2023）。 - 删除练习试次的数据。 - 删除指导语试次或试次中非选择环节的数据。 extractedData.csv 包含以下数据列： - subjID_anon（整数型，匿名化被试ID，取值范围1至279） - rt（整数型，反应时，单位为毫秒） - response（字符串型，被试的键盘作答结果，仅包含"arrowleft"与"arrowright"） - category（字符串型，当前行的类别数据，本研究中仅选定类别"choice"） - success（整数型，二分类指标，表示主动行为是否成功：成功为1，损失或被动行为为0；未作出反应的被试标记为"noResponse"） - robberTotPoints（整数型，当前抢劫任务区块内累计获得的点数；工厂任务试次该字段为0，且每个区块起始点数均为0） - factoryTotPoints（整数型，当前工厂任务区块内累计获得的点数；抢劫任务试次该字段为0，且每个区块起始点数均为0） - trial（整数型，任务内试次编号：第一区块为0至59，第二区块为60至119） - task（字符串型，当前试次所属的任务类型：抢劫任务为"robber"，工厂任务为"factory"） - offers（整数型，向被试呈现的任务提示信息：在抢劫任务中，该值为被试在遭遇抢劫前的夜间可获得的小费金额；在工厂任务中，该值为被试放任机器损坏时可能承受的损失金额） - cost（整数型，当前试次中主动行为的成本） - noActValue（整数型，被动行为的价值：抢劫任务中为0，工厂任务中为-offers） - stimNo（字符串型，当前试次呈现的刺激文件名称：抢劫任务示例为"img/robber1.png"，工厂任务示例为"img/factory1.png"） - probWin（浮点型，当前试次使用的成功概率，基于二项分布计算，用于判断被试选择主动行为时能否成功避免负面结果） - act（整数型，二分类指标，表示被试的选择：主动行为为1，被动行为为0；未作出反应的被试标记为"noResponse"） - PosAffect_tot（整数型，被试在积极与消极情感量表（PANAS）积极情感子量表上的总分（Watson等人，1988）） - STAIT_tot（整数型，被试在状态-特质焦虑量表（STAI）特质焦虑子量表（Y型）上的总分（Spielberger等人，1983）） - demograph_age（整数型，提取自Prolific平台的人口统计学数据中的"Age"字段，单位为年） - demograph_sex（字符串型，提取自Prolific平台的人口统计学数据中的"Sex"字段，男性为"Male"，女性为"Female"） - demograph_nationality（字符串型，提取自Prolific平台的人口统计学数据中的"Nationality"字段） - demograph_approvals（整数型，提取自Prolific平台的人口统计学数据中的"Total.approvals"字段） - failed_AC_Q（整数型，被试在问卷中未通过注意力检查的次数（0至6），若未通过次数超过2次则该被试被排除） extractedData_FU.csv 包含与上述数据集相同的列，仅移除了"demograph_approvals"字段，替换为： - demograph_dateDif（整数型，两次测试会话的间隔天数，通过提取自Prolific平台的人口统计学数据中两次会话的"Completed.at"字段计算得出，即第二次会话完成时间减去第一次会话完成时间） 本实验任务由卡罗林斯卡学院的托比亚斯·格兰瓦尔德开发，采用JavaScript、HTML与CSS编写以支持线上数据收集，同时使用了jsPsych v7.1.0工具包（de Leeuw, 2015）。 本实验任务可通过以下链接获取：https://github.com/Granwald/healthyPriors ## 参考文献 1. de Leeuw, J. R. (2015). jsPsych: A JavaScript library for creating behavioral experiments in a Web browser. *Behavior Research Methods*, 47(1), 1–12. https://doi.org/10.3758/s13428-014-0458-y 德·勒乌，J.R.（2015）。jsPsych：用于在网页浏览器中创建行为实验的JavaScript库。《Behavior Research Methods》，47(1)，1-12。https://doi.org/10.3758/s13428-014-0458-y 2. Granwald, T., Dayan, P., Lengyel, M., & Guitart-Masip, M. (2025). A Task-Invariant Prior Explains Trial-by-Trial Active Avoidance Behaviour Across Gain and Loss Tasks. *Communications Psychology*, 82(3) 1-14. https://doi.org/10.1038/s44271-025-00254-1 格兰瓦尔德，T.，戴扬，P.，伦耶尔，M.，& 吉塔特-马西普，M.（2025）。一项任务不变先验解释跨收益与损失任务的逐试次主动回避行为。《Communications Psychology》，82(3)，1-14。https://doi.org/10.1038/s44271-025-00254-1 3. Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). State-Trait Anxiety Inventory for Adults Sampler Set Manual, Instrument and Scoring Guide. www.mindgarden.com. 施皮尔伯格，C.D.，戈沙奇，R.L.，卢申，R.，瓦格，P.R.，& 雅各布斯，G.A.（1983）。《成人状态-特质焦虑量表套装手册、工具与计分指南》。www.mindgarden.com。 4. Stark, C. E. L., Noche, J. A., Ebersberger, J. R., Mayer, L., & Stark, S. M. (2023). Optimizing the mnemonic similarity task for efficient, widespread use. *Frontiers in Behavioral Neuroscience*, 17. https://www.frontiersin.org/articles/10.3389/fnbeh.2023.1080366 斯塔克，C.E.L.，诺切，J.A.，埃伯斯伯格，J.R.，迈耶，L.，& 斯塔克，S.M.（2023）。优化记忆相似性任务以实现高效广泛应用。《Frontiers in Behavioral Neuroscience》，17。https://www.frontiersin.org/articles/10.3389/fnbeh.2023.1080366 5. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and Validation of Brief Measures of Positive and Negative Affect: The PANAS Scales. *Journal of Personality and Social Psychology*, 54(6), 1063–1070. https://doi.org/10.1037/0022-3514.54.6.1063 沃森，D.，克拉克，L.A.，& 特勒根，A.（1988）。积极与消极情感简版量表的编制与验证：PANAS量表。《Journal of Personality and Social Psychology》，54(6)，1063-1070。https://doi.org/10.1037/0022-3514.54.6.1063