Research Article: Elaboration versus Suppression of Cued Memories: Influence of Memory Recall Instruction and Success on Parietal Lobe, Default Network, and Hippocampal Activity

Date Published: February 19, 2014

Publisher: Public Library of Science

Author(s): Sarah I. Gimbel, James B. Brewer, Bart Rypma.


Functional imaging studies of episodic memory retrieval consistently report task-evoked and memory-related activity in the medial temporal lobe, default network and parietal lobe subregions. Associated components of memory retrieval, such as attention-shifts, search, retrieval success, and post-retrieval processing also influence regional activity, but these influences remain ill-defined. To better understand how top-down control affects the neural bases of memory retrieval, we examined how regional activity responses were modulated by task goals during recall success or failure. Specifically, activity was examined during memory suppression, recall, and elaborative recall of paired-associates. Parietal lobe was subdivided into dorsal (BA 7), posterior ventral (BA 39), and anterior ventral (BA 40) regions, which were investigated separately to examine hypothesized distinctions in sub-regional functional responses related to differential attention-to-memory and memory strength. Top-down suppression of recall abolished memory strength effects in BA 39, which showed a task-negative response, and BA 40, which showed a task-positive response. The task-negative response in default network showed greater negatively-deflected signal for forgotten pairs when task goals required recall. Hippocampal activity was task-positive and was influenced by memory strength only when task goals required recall. As in previous studies, we show a memory strength effect in parietal lobe and hippocampus, but we show that this effect is top-down controlled and sensitive to whether the subject is trying to suppress or retrieve a memory. These regions are all implicated in memory recall, but their individual activity patterns show distinct memory-strength-related responses when task goals are varied. In parietal lobe, default network, and hippocampus, top-down control can override the commonly identified effects of memory strength.

Partial Text

The retrieval of episodic memories is elemental to nearly all aspects of everyday life, yet little is known about how the brain performs and integrates the component processes of memory retrieval. Some brain functions might be called upon specifically for episodic memory retrieval, while other functions, also perhaps critical for memory retrieval, are broadly involved in a range of cognitive processes. Imaging and electrophysiological studies have sought to link regional brain activity to episodic memory function and have demonstrated that activity in the human hippocampus is modulated during both encoding and retrieval of memories [1]. Further, these studies have provided evidence supporting top-down modulation of hippocampal activity during memory retrieval and suppression. The role of the parietal lobe in episodic memory retrieval is controversial, and interpretation is complicated by the fact that this region includes several subregions with distinct functions, functional responses, and connectivity [2]. Electrophysiological studies suggest that regions of the parietal lobe are engaged prior to recall of a memory [3], supporting that these regions are driven by direction of attention toward memory [4]–[6]. The present study examines how task goals modulate regional activity during memory recall and additionally modulate memory strength effects. Given what is already known about regional contributions to these tasks, the analysis will focus on subregions of the parietal lobe, medial temporal lobe, and default network regions. By identifying how recall-related brain activity is influenced by task goals, or top-down processes, this work is a step toward understanding how task instruction and recall success modulate activity in human medial temporal lobe, parietal lobe and default network regions.

This experiment used studied pairs presented under differing retrieval conditions to explore how parietal lobe subregions (Figure 2), default network (Figure 3), and posterior hippocampus (Figure 4) might be influenced by top-down processes directing attention to memory during retrieval success and failure. The findings suggest that the dorsal parietal region (BA 7) is minimally responsive to success and failure, but is greatly influenced by attention to memory. In contrast, ventral regions of the parietal lobe (BA 40 and 39) are highly sensitive to retrieval success, but only under conditions where retrieval is attempted. Directed suppression of retrieval abolishes any retrieval success effect in all three of the parietal regions. Default network was influenced by retrieval success, showing a larger decrease in activity for trials that were forgotten. Hippocampus was sensitive to retrieval success and attention to memory, suggesting that regional hippocampal activity can be influenced by top down processes directing attention toward or away from retrieval.

By examining top-down influences on parietal and hippocampal brain activity linked to memory retrieval, this study highlights how attention-shifts, search, and post-retrieval processing are important drivers of such activity, in some cases even more than the retrieval event itself. This study found that top-down suppression affects early stimulus processing, possibly gating bottom-up capture of attention. Additionally, dorsal parietal cortex was found to be responsive to top-down attention to memory. This study adds to the body of literature suggesting functional heterogeneity of parietal lobe subregions during memory retrieval. Memory strength effects were found in the default network, and it was revealed that the primary modulator of these effects was further suppression of activity for forgotten responses. It was also found that hippocampal retrieval-success effects are dependent on the top-down direction of attention toward or away from memory retrieval. In this study, parietal lobe regions and hippocampus were defined anatomically. Finer-grained delineation within accepted anatomical regions could yield further functional dissociation even within a defined region [49]–[51]. Studying the interplay and dissociation of “ancillary” memory processes that include attention-shifts, search, and post-retrieval processing will provide a more complete explanation of the regional contributions to oft-described patterns of neuronal activity seen during memory retrieval.