Effect of fornix stimulation on acute memory encoding

The reported memory-related effects of stimulating different nodes within the classical Circuit of Papez have been inconsistent, though the possibility of immediate or chronic memory benefit remains. We undertook memory experiments during deep brain stimulation (DBS) in human subjects with mild, early Alzheimer's disease (2 males, 3 females; n = 5) to rigorously investigate the effects of fornix stimulation on acute memory function. We sought to assess whether some of the reported variability may be attributed to the stimulation protocol and to determine whether stimulation alters memory formation per se, rather than influencing a process required for but not directly related to memory encoding. We tested fornix stimulation during both awake DBS surgery and postoperatively, in subjects participating in a broader clinical trial (ADvance II). In both settings, using a parametric episodic memory task with distinct encoding, attention, and delayed recall phases, we found that fornix stimulation generally impaired memory, with higher frequencies producing the greatest detriments on memory performance. Furthermore, stimulation specifically interacted with trial-by-trial memory encoding, rather than with other functions such as visual-spatial processing, attention, or short-term working memory. Therefore, in both contexts and across a wide range of stimulation frequencies, open-loop fornix stimulation directly impaired acute, item-specific memory encoding, though the effects of more chronic stimulation on memory and cognitive function are yet to be determined.NEW & NOTEWORTHY Electrical stimulation of different nodes within the Circuit of Papez has been assessed in a variety of human studies with conflicting or indeterminate results. Our goal was to combine a precise psychophysical paradigm with varying stimulation protocols across two experimental settings (intra-op, post-op) to more robustly identify and characterize the interaction between fornix stimulation and memory and to isolate whether those effects are specific to memory encoding, rather than related to distinct, supporting cognitive functions.