Self-reports map the landscape of task states derived from brain imaging.
Description
Data for Mckeown, Goodall-Halliwell et al. (2024) "Self-reports map the landscape of task states derived from brain imaging." Summary: Psychological states influence our happiness and productivity; however, estimates of their impact have historically been assumed to be limited by the accuracy with which introspection can quantify them. Over the last two decades, studies have shown that introspective descriptions of psychological states correlate with objective indicators of cognition, including task performance and metrics of brain function, using techniques like functional magnetic resonance imaging. Such evidence suggests it may be possible to quantify the mapping between self-reports of experience and ‘ground-truth’ or biologically related representations of those states (e.g., those inferred from measures of brain activity). Here, we used machine learning to show that self-reported descriptions of experiences across tasks can reliably map the objective landscape of task states derived from brain activity. In our study, participants provided descriptions of their psychological states while performing tasks for which the contribution of different brain systems was available from prior fMRI studies. We used machine learning to combine these reports with descriptions of brain function to form a ‘state-space’ that reliably predicted patterns of brain activity based solely on unseen descriptions of experience. Our study demonstrates that introspection contains information that maps the objective task landscape inferred from brain activity. Original Sample: 194 participants were recruited to complete the full 14-task battery in a behavioural laboratory. Demographic information was missing for four participants due to errors in data collection. These four participants were excluded from all analyses, resulting in a final sample of 190 participants. Of these 190 participants, 164 identified as women, 24 as men, and two as non-binary or similar gender identity. Mean age of participants was 18.56 years (SD = 1.09, range = 17 to 24 years). All 190 participants contributed 38 observations each, resulting in 7220 total observations. Replication: 101 participants were recruited to complete a subset of 4 tasks from the full 14-task battery. Demographic information was missing for five participants due to errors in data collection. These five participants were excluded from all analyses. In addition, one participant was removed from analyses as they were missing over half their data due to a technical error during data collection, leaving a final sample of 95 participants. Of the remaining 95 participants, 87 identified as women, 6 as men, and 2 as non-binary or similar gender identity. Mean age of participants was 18.24 years (SD = 1.05, range = 17 to 24 years). 94 out of 95 participants contributed 11 observations each, with one participant missing one observation due to a technical error in data collection, resulting in 1044 total observations.
Files
Steps to reproduce
Full 14-Task Battery: Participants attended one 2-hour study session in the behavioral laboratory to complete the 14-task battery. Participants provided written informed consent before the study session began and provided their demographic information via a demographic questionnaire. Once consent and demographic information were acquired, participants completed the 14-task battery in a room alone with a computer. Participants were asked to refrain from using any technological devices (phones, smartwatches, tablets, laptops/computers) besides the computer in front of them to avoid distractions. The task battery was presented using PsychoPy35 and included the following 14 tasks, grouped into seven pairs based on task similarity: 1) Easy-/Hard-Math, 2) Finger-Tapping and Go/No-Go, 3) Self-/Friend-Reference, 4) 0-/1-Back, 5) 2-Back-Faces/-Scenes, 6) Autobiographical Memory and Reading, and 7) passive viewing of Documentary and Sci-Fi videos. There were three task blocks per task, except for the passive-viewing (Documentary and Sci-Fi) and Two-Back (Faces and Scenes) tasks, which had two blocks each (38 task blocks total). Within each task, task blocks were randomized (except the passive-viewing tasks). Across tasks, each task block lasted ~90s, jittered by +/- 15s. Task pairs were presented consecutively. Within task pairs, task order presentation was randomized, and the task pairs were randomized based on a unique seed generated for each participant. Written instructions were presented at the start of each task block. Immediately after each task block, participants were prompted with mDES questions about their thoughts during the previous block. The task battery took participants approximately 1.5-2 hours to complete. Full code for the presentation of the 14-task battery is openly available on GitHub: https://github.com/ThinCLabQueens/Task-Battery/tree/production. Shortened 4-Task Battery: The procedural details for the shortened task battery were identical to those of the full task battery, except only four tasks were presented in a randomized order across participants: 1) Hard-Math, 2) Go/No-Go, 3) Autobiographical Memory 4) Documentary. There were three task blocks per task, except for the passive-viewing Documentary task which had two blocks (11 task blocks total). The task battery took participants approximately 30-60 minutes to complete. Full code for the presentation of the 4-task battery is openly available on GitHub: https://github.com/ThinCLabQueens/Shortened-Task-Battery.