Background. Our previous studies have implicated the primary visual cortex (V1) as the putative visuo-spatial 'sketchpad' for working-memory, but in a supramodal form. To establish this memory-related role for V1, we need to determine the source of its top-down modulation from higher-order memory mechanisms, including medial-temporal lobe (MTL) structures such as the hippocampus and perirhinal cortex (PRC) (Likova, 2012, 2013), which has direct anatomical connection to V1 (Clavagnier et al., 2004). Indeed, V1 and the hippocampus exhibited correlated changes under a memory-based training intervention (Likova, 2015); moreover, the representations for both memory and perception were confirmed as supramodal in PRC (Cacciamani & Likova, 2016). Now, to address the key question of the direction and significance of influence between these memory areas and V1, we ran Granger Causality analysis. Methods. Using fMRI in blind subjects before and after a unique memory-guided drawing intervention (Cognitive-Kinesthetic training), previously shown to generate V1 reorganization towards tactile-memory function in the blind, we ran three tasks (20s each): tactile exploration of raised-line drawings of faces and objects, tactile memory retrieval via drawing, and a scribble motor/memory control. Results. Comparative pre/post Granger Causality analysis revealed a significant increase in hippocampus-to-V1 and PRC-to-V1 causal influence after training with the memory-drawing task, but not during the control task, indicating that the drawing-from-memory training strengthened the top-down effect on visual cortex from these MTL structures. Conclusion. This is the first study to demonstrate causal connectivity from the hippocampus and PRC to V1. That this happens as a result of a memory-training intervention supports our hypothesis of the role of these memory structures as a top-down source for the cortical reorganization of V1 in the blind, and is consistent with its proposed function as a supramodal working-memory 'sketchpad' for the active processing of detailed spatial information (Likova, 2012).
Publication Type: Presentation
Authors: Lora T Likova; Laura Cacciamani; S.C. Nicholas; Kristyo N Mineff
PMCID/PMID: J Vision 17
DOI URL: http://dx.doi.org/10.1167/17.10.594