Causal necessity of hippocampus for structure-based inference in learning

Abstract

When meeting new individuals or encountering known individuals in new circumstances, we intuitively map out their relationships – not merely by direct experience, but by quickly inferring new connections based on prior relational knowledge. Using a novel task, we demonstrated that participants indeed employ knowledge of relational structures to facilitate learning of new relationships in a changing environment. Computational modelling revealed that participants leveraged relational knowledge to support inference, thus facilitating learning. Whole brain neuroimaging identified a uniquely robust representation of relational structure in the hippocampus. Neural networks trained on similar tasks demonstrated the emergence of relational structure representations, resembling those found in hippocampus. Lesioning network units sustaining such representations disrupted structure-based inference and predicted hippocampus’s essential role. Transcranial ultrasound stimulation of human hippocampus, transiently modulating its activity without affecting overlying tissue, produced similar disruption effects, empirically confirming the causal necessity of hippocampal representations for structure-based inference in learning.

Abstract via europepmc.

Flags from extraction

• exposures[0].timing.pulse_repetition_frequency_hz — 5 Hz stated in Results as theta-burst (5 Hz, 20-ms pulses every 200 ms); the Methods give pulse repetition interval 200 ms.
• exposures[0].timing.waveform — Paper calls the protocol theta-burst patterned TUS, but it is a single 5 Hz train of 20 ms pulses, with no nested burst structure.
• exposures[0].in_situ.isppa_w_cm2 — K-Plan simulation, approximately 5 W/cm2 on average across participants (min 3 W/cm2 required for inclusion); value read from text, spread in a figure only.
• exposures[0].free_field.isppa_w_cm2 — 75 W/cm2 is for hippocampal stimulation; the white matter control used 55 to 75 W/cm2.
• exposures[0].device.family — TPO named, so neurofus; transducer is the four-element DPX-500-4CH.
• sham_type — Control is real ultrasound to a hippocampus-unrelated white matter site, intensity-matched in situ; no inactive-transducer sham.
• randomised — Session order was counterbalanced; randomisation not stated.
• direction_of_effect — TUS to hippocampus disrupted structure-based inference relative to white matter control; authors suggest noise induction rather than clear excitation or inhibition.
• adverse_events — A post-TUS questionnaire was given (immediately and at 24 h) but its results are not reported in the main text.
• auditory_control — No auditory control or discussion of auditory confounds found.