Mapping the metabolic correlates of working memory in a naturalistic environment: a whole-room indirect calorimetry study

Abstract Whole-room indirect calorimetry (WRIC) is an advanced platform to assess basal and activity-related metabolic expenditure, but its sensitivity to cognitive activity-induced metabolic fluctuations remains largely unexplored. In this exploratory study, we investigated whether WRIC measurements covary with variations in working memory load. Eighteen healthy adults completed two experimental sessions inside a WRIC chamber: a working memory task (n-back, with increasing difficulty) and a control simple reaction time (SRT) task. Oxygen consumption and carbon dioxide production were continuously recorded to derive metabolic rate (MR) and respiratory exchange ratio (RER) at 1-min resolution, alongside subjective workload ratings (NASA-TLX) and electrodermal activity for autonomic arousal. Linear mixed-effects models applied to 4-min downsampled WRIC data revealed significant effects of task session, block, and subjective workload on both MR and RER, even after controlling for autonomic arousal. Downsampling reduced residual serial dependence while preserving the time-varying structure of the physiological predictors. Block-wise metabolic fluctuations were observed across the cognitive and control sessions, although nonlinear patterns at high task difficulty should be interpreted cautiously in the absence of direct behavioral evidence for performance floor effects. These findings provide preliminary evidence that WRIC-derived metabolic measures may provide an ecologically valid exploratory approach for investigating cognition-metabolism interactions.