Research ReportEncoding of visual–spatial information in working memory requires more cerebral efforts than retrieval: Evidence from an EEG and virtual reality study
Introduction
Navigating through space requires a variety of cortical networks. These networks are specialized depending on the type of question being asked. For example, in perception, a distinction can be made between the spatial component and the non-spatial component resulting in two separate visual processing pathways asking the questions of ‘what is it? (object type) and where is it? (object spatial location) (Smith, 1995, Wilson et al., 1993). Neuroimaging studies support the formulation that spatial processing chiefly involves dorsal stream—occipitoparietal pathway, whereas object processing tend to involve ventral, occipitotemporal pathway (Ungerleider and Mishkin, 1982, Ungerleider and Haxby, 1994). This specificity has been found to extend to the prefrontal cortex, with dorsolateral area responsible for spatial working memory and ventrolateral area responsible for nonspatial working memory (Goldman-Rakic, 1987, Courtney et al., 1996, Levy and Goldman-Rakic, 1999). Thus, there is a consensus that most working memory tasks recruit a network of dorsolateral prefrontal cortex and parietal and occipital areas (Jonides et al., 1993). It has also been well documented that brain lesions, cerebral ischemia, aging, and pharmacological agents can result in disturbance in these visual–spatial networks (Miyamoto et al., 1987, McNamara and Skelton, 1993) and that deficits in spatial working memory result from dysfunction involving the right posterior parietal and right dorsolateral prefrontal cortex (van Asselen et al., 2006).
Spatial navigation also incorporates a variety of cortical systems. Both behavioral and neuroimaging studies suggest that distinct brain areas become involved according to the nature of the task (Hartley et al., 2003, Janzen and Weststeijn, 2007, Voermans et al., 2004). Utilizing research suggesting two distinct memory systems in navigation, one for route direction and one for relevant object location, Janzen and Weststeijn (2007) suggested that the inferior parietal gyrus, the anterior cingulated gyrus and the right caudate nucleus are involved in coding of route direction whereas the parahippocampal gyri distinguish between relevant and irrelevant landmarks. A similar distinction based on both human and animal data suggest that the hippocampus is involved in acquiring a cognitive map of the environment, whereas the caudate nucleus is involved in learning place appropriate responses leading to habitual behavior (Voermans et al. 2004). Another similar distinction has been made between wayfinding in humans which involves adopting a novel route and route following which involves following a previously used route suggests differential involvement of the hippocampus for the first and the caudate for the second (Hartley et al. 2003).
To study spatial memory and navigation, we created a virtual reality (VR) environment (virtual corridor) in which individuals encoded spatial information by being shown the navigation route to the target room and later retrieved the internal information necessary to navigate through the corridor and reproduce the exact route to the target room. As noted in a recent fMRI study, few neuroimaging studies have utilized designs which have allowed for a comparison between encoding and retrieval (Kim et al., 2010).
One distinction between encoding and retrieval is that encoding emphasizes the processing of external stimuli whereas retrieval involves internal processes. This distinction between externally directed and internally generated processing has produced a large number of studies referred to as baseline or default network as well as referred to as task positive and task negative. As overviewed by Kim et al., the task positive network includes lateral prefrontal cortex (PFC), dorsal parietal cortex, sensory–motor cortices, subcortical areas, and the cerebellum (Cabeza and Nyberg, 2000, Naghavi and Nyberg, 2005, Shulman et al., 1997a), whereas the task-negative network, also known as the default-mode network (Raichle et al., 2001), consists of anterior and medial PFC, the precuneus, and the angular gyrus (Shulman et al., 1997b, Gusnard and Raichle, 2001). A recent fMRI study has shown differential involvement of the default cortical network in terms of encoding and retrieval (Daselaar et al., 2009). In addition, activity in these areas has also been associated with successful retrieval of information (Wagner et al., 2005). Increased activity in PMR and VPC during successful retrieval and reduced activity during successful encoding has been reported (Daselaar et al., 2009). Given that frontal EEG theta may be viewed as an index of default network activity (Scheeringa et al., 2008), we would expect to find differential encoding/retrieval differences in EEG theta activity.
In the present study, we sought to examine spatial processing during encoding and retrieval of a spatial memory task. Specifically, we designed an EEG study that incorporates spatial processing using a virtual reality (VR) environment. We developed VR-based spatial memory tasks including encoding and retrieval conditions enabling the subjects (a) to experience the sense of presence (Jancke et al., 2009) and (b) to track the changing brain activation patterns during exposure to these VR-driven spatial memory task conditions. It was hypothesized that encoding of visual–spatial information in working memory requires more cerebral efforts in comparison to retrieval and that these processes differentially activate the default network.
Section snippets
Behavioral results
The success rate of the task performance, i.e., accurate navigation to the target room was 100%. The average time taken by complete the entire task, from the start room to target room and back to start room was 52.58 ± 6.23 seconds. The average time taken by subjects to go from start room to target room was 24.83 ± 3.23 seconds, and the average time to come back from target room to start room was found to be 27.75 ± 2.87 seconds. A paired t-test was used to see if the difference was significant. T =
Discussion
Our results supported the hypothesis that spatial encoding required more effort than retrieval and that these processes differentially involve the default network. This was shown by the increase in theta power from baseline as compared to encoding, whereas there was a decrease in theta power during retrieval, suggesting that encoding is more demanding or cognitively challenging compared to baseline or retrieval. Frontal–central predominance was seen during encoding when compared with baseline
Conclusion
Overall, it can be concluded from our EEG power analysis and source localization using sLORETA that encoding requires more cerebral efforts from the individual in comparison to retrieval of information during a visual–spatial working memory task. Our results are in support of encoding/retrieval flip, as different neural substrates were employed during encoding and retrieval of information. Also, the pattern of brain areas activated during both encoding and retrieval supports the concept of
Participants and study design
Twelve neurologically normal students were recruited from The Pennsylvania State University for this experiment. All the subjects were right-handed according to Edinburgh Handedness Inventory (Oldfield, 1971) and had normal or corrected-to-normal vision. The subject's included seven male and five female subjects with an average age of 22.67 years. They signed an informed consent to a protocol approved by the Institutional Review Board of The Pennsylvania State University. The participants
Acknowledgments
This study was supported by NIH grant RO1 NS056227-01A2 “Identification of Athletes at Risk for Traumatic Brain Injury” awarded to Dr. Slobounov, PI. We would like to thank Elena Slobounov for VR programming.
References (42)
- et al.
Sources of cortical rhythms in adults during physiological aging: a multi-center EEG study
Clin. Neurophysiol.
(2006) - et al.
Resting state cortical rhythms in athletes: a high-resolution EEG study
Brain Res. Bull.
(2010) - et al.
The right parietal lobe is critical for visual working memory
Neuropsychologia
(2008) - et al.
EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis
J Neurosci Methods.
(2004) - et al.
Maintenance versus manipulation of information held in working memory: an event-related fMRI study
Brain Cogn.
(1999) - et al.
The well-worn route and the path less traveled: distinct neural bases of route following and wayfinding in humans
Neuron
(2003) - et al.
Neural representation of object location and route direction: an event-related fMRI study
Brain Res.
(2007) - et al.
Overlapping brain activity between episodic memory encoding and retrieval: roles of the task-positive and task-negative networks
Neuroimage
(2010) - et al.
Theta synchronization during episodic retrieval: neural correlates of conscious awareness
Brain Res. Cogn. Brain Res.
(2001) - et al.
The effects of memory load on event-related EEG desynchronization and synchronization
Clin. Neurophysiol.
(2000)