Distributed and Dynamic Storage of Working Memory Stimulus Information in Extrastriate Cortex

Abstract

The predominant neurobiological model of working memory (WM) posits that stimulus information is stored via stable elevated activity within highly selective neurons. Based on this model, which we refer to as the canonical model, the storage of stimulus information is largely associated with lateral prefrontal cortex (lPFC). A growing number of studies describe results that cannot be fully explained by the canonical model, suggesting that it is in need of revision. In the present study, we directly test key elements of the canonical model. We analyzed functional MRI data collected as participants performed a task requiring WM for faces and scenes. Multivariate decoding procedures identified patterns of activity containing information about the items maintained in WM (faces, scenes, or both). While information about WM items was identified in extrastriate visual cortex (EC) and lPFC, only EC exhibited a pattern of results consistent with a sensory representation. Information in both regions persisted even in the absence of elevated activity, suggesting that elevated population activity may not represent the storage of information in WM. Additionally, we observed that WM information was distributed across EC neural populations that exhibited a broad range of selectivity for the WM items rather than restricted to highly selective EC populations. Finally, we determined that activity patterns coding for WM information were not stable, but instead varied over the course of a trial, indicating that the neural code for WM information is dynamic rather than static. Together, these findings challenge the canonical model of WM.

Protracted development of executive and mnemonic brain systems underlying working memory in adolescence: A longitudinal fMRI study

“…This finding is consistent with the Sensorimotor Hypothesis of WM [37] that highlights the role of visual cortex in providing optimal storage of visuospatial information while prefrontal regions support the executive aspect of generating a voluntary response. This proposal is supported by studies using multivariate analyses of fMRI data showing that the stimulus characteristics stored in WM are encoded exclusively in visual cortex [38, 39, 40, 41]. Importantly, activity in visual cortical regions during encoding was associated with accuracy across development (after controlling for age), suggesting that developmental improvements in WM are predominantly underlied by enhanced integration of systems specific to the task, namely visuospatial processing.…”

Parietal and Frontal Cortex Encode Stimulus-Specific Mnemonic Representations during Visual Working Memory

“…For example, several studies indicate that while specific features of a remembered stimulus (e.g., motion direction, orientation, color) can be successfully decoded using activation in posterior visual areas that do not show an elevated mean response during WM, feature information cannot not be decoded in subregions of frontoparietal cortex that do show an elevated mean response during WM (Riggall and Postle, 2012; Emrich et al, 2013; Lee et al, 2013; Sreenivasan et al, 2014b). To examine whether feature-selective representations might also be present in these frontoparietal areas, we next examined multivoxel activation patterns related to remembered and non-remembered orientations in cortical areas with elevated delay period activation.…”

Interference and memory capacity limitations.

“…However, at least for the within-vision stores, an alternative interpretation is that items from different categories show less interference than items from the same category, especially if, as recent brain imaging studies suggest, memory items are stored in those brain regions where they are processed to begin with (e.g., Lee, Kravitz, & Baker, 2013;Riggall & Postle, 2012;Sreenivasan, Vytlacil, & D'Esposito, 2014;Sreenivasan, Curtis, & D'Esposito, 2014):…”