Fig. 1. Button presses indicate the timing of memory-dependent and -independent processing.

A Structure of the memory task. Two groups of participants (groups 1 and 2; n = 226) completed blocks consisting of an encoding phase (top row) in which they associated cues (“spin”, “cut”) to objects, a distractor phase (not shown), and one of two versions of a retrieval phase (bottom rows), in which they answered catch questions about remembered objects (“animate or inanimate?”, “photo or drawing?”). B Structure of the visual task. A separate group of participants (group 3; n = 95) answered questions about objects on the screen, using the same questions and stimuli as the memory task. C Number of participants that were included in further analyses, after exclusion of participants with a high number of incorrect and/or timed-out trials (Supplementary Fig. 1A). Note that participants in group 1 contributed button presses to three task phases (Encoding, Retrieval, and Catch-after-retrieval), group 2 contributed to 2 (Encoding and Catch-with-retrieval), and group 3 to 1 task phase (Visual). The example stimuli shown in A and B were taken from the BOSS database⁷⁰.