The role of precuneus and left inferior frontal cortex during source memory episodic retrieval
Introduction
Human memory that is important for encoding and retrieving declarative information has been fractionated into working, semantic, and episodic memory (e.g., Eichenbaum and Cohen, 2001, Schacter and Tulving, 1994). Roughly speaking, semantic memory comprises general world knowledge, such as word meaning and word use, and it functions to store, retrieve, and associate this information to environmental stimuli. On the other hand, working memory refers to the short-term maintenance and manipulation of information during processing. Finally, episodic memory is employed for long-term storage and recall of previous experiences or episodes, and it allows people to reflect upon their personal past. Entailing more than just event memory and event recall, some have suggested that episodic memory necessarily includes a special awareness for subjective time known as autonoetic consciousness; it enables people to mentally travel backwards in time and knowingly retrieve information from a given personal experience (e.g., Baddeley, 2001, Tulving, 2002, Wheeler, 2000). In other words, episodic memory allows current knowledge to be associated with past experiences.
Source memory tasks have commonly been employed to study episodic memory in neuroimaging (e.g., Rugg and Henson, 2002). However, it has been noted that many processes are likely active during memory retrieval, and several recent studies have emphasized the importance of parsing the effects of these processes from each other (Dobbins et al., 2003, Kahn et al., 2004, Rugg et al., 2003). For example, recognition memory can be thought of as comprising at least two processes, recollection and familiarity (e.g., Kelley and Jacoby, 2000, Mandler, 1980, Yonelinas et al., 1996). Recollection involves a detailed recognition that helps associate facts with personal experience, while familiarity entails only a sense of resemblance. Retrieval can also be operationally defined as encompassing multiple processes, such as mode, orientation, success, and effort (Rugg and Wilding, 2000). Retrieval mode is a relatively long-standing state relating to the task at hand, while orientation is a subset of mode that helps determine how a specific retrieval cue, sometimes known as a retrieval probe, will be processed. For example, cues within a task that elicit different kinds of correct answers would then have different orientations, such as between item recognition vs. source memory judgments or cues relating to previously encoded pictures vs. words. Finally, retrieval success relates to the recovery of previously encoded information, while effort relates to task difficulty and is usually assessed by accuracy or reaction times.
From a previous study of source memory and item recognition (Lundstrom et al., 2003), data suggest that the retrieval of contextual associations, specifically an imagined picture paired with an encoded word, is associated with activation in the posterior precuneus and left prefrontal cortex. Our explanation for the data has been that successful source memory requires the two-step process of context revival and subsequent processing, where the posterior precuneus activates during regeneration of previous contextual associations and the left ventro-lateral prefrontal cortex (PFC) mediates explicit retrieval and integration of the contextual associations (Lundstrom et al., 2003). This is in accord with previous studies of episodic memory retrieval showing precuneual activation (e.g., Cabeza and Nyberg, 2000, Rugg and Henson, 2003) as well as work that has shown the precuneus to be involved in mental imagery recall (e.g., Fletcher et al., 1995, Grasby et al., 1993, Shallice et al., 1994) and retrieval independent of imagery (Krause et al., 1999). Further, neuroimaging studies have noted left-lateralized activation of the PFC and learning-related decreases, i.e., decreased retrieval activation after increased encoding practice, during source memory retrieval tasks (Dobbins et al., 2002, Nolde et al., 1998a, Petersson et al., 2001). However, in our previous study, the left PFC and posterior precuneus were activated in tandem, and it is difficult to determine the specific roles the two areas may play.
In this study, we use the same source memory paradigm as in our previous study (Lundstrom et al., 2003) but increase the latency between encoding and retrieval from ∼1 min to ∼3 days. This increases subject error rates and allows us to make comparisons (e.g., correct source vs. miss and miss vs. new trials) that should, for example, isolate retrieval success from retrieval orientation. From previous experience with this paradigm, we know that subjects effectively encode source memory associations and can subsequently retrieve these associations within minutes with ∼88% accuracy (Lundstrom et al., 2003). Lower accuracy with increased latency is then expected to represent a failure to regenerate and/or retrieve contextual associations rather than a failure to encode them.
The primary purpose of this study is to further examine posterior precuneal and left prefrontal activation during episodic source memory retrieval in order to suggest more specific roles for the posterior precuneus and left PFC. Specifically, we expect activation of the posterior precuneus during comparisons where successful retrieval of relevant information occurs, such as during correct source minus false alarms (as predicted by Gonsalves and Paller, 2000). We also expect that areas in the left ventro-lateral inferior region/frontal operculum will be active when retrieval and integration of information is attempted, such as during correct source, incorrect source, and miss trials. Data from a previous study (Lundstrom et al., 2003) suggested a functional segregation of the left PFC, and we also noted activation of the left ventro-lateral frontal region/frontal operculum during a comparison of source memory old/new effects vs. item recognition old/new effects, suggesting that the area is involved in a process that was more active during source memory than item recognition retrieval, such as information searching and integration. This area has been noted in previous neuroimaging studies of episodic and source memory (e.g., Buckner et al., 1998, Rugg et al., 1999, Takahashi et al., 2002). Similarly, a recent study of world knowledge as well as lexical–semantic integration implicates a left inferior frontal region in the vicinity of Brodmann area 47 (Hagoort et al., 2004).
Section snippets
Participants
Sixteen healthy, right-handed volunteers (13 female) with an average age of 24 years (range 20–30, SD = 2.8) participated in the study. All were pre-screened, had no history of previous neurological conditions, and gave written informed consent. The ethics committee of Karolinska Hospital/Institutet approved the study.
Experimental stimuli and design
120 common concrete nouns were divided into three 40-word groups, which were matched for word length (means 4.2–5.1 letters, range 3–8 letters) and frequency (means 56.3–60.0 per
Behavioral performance
Overall, subjects displayed accuracies that were significantly greater than chance and reaction times that were not significantly different from each other (Table 1). Given random responses, accuracies within each of the columns of Table 1A (“Viewed”, “Imagined”, and “New”) would be approximately equal; however, this is not the case for column 1 “Viewed” [F(2,45) = 56.3, P < 0.001], column 2 “Imagined” [F(2,45) = 63.7, P < 0.001], or column 3 “New” [F(2,45) = 157.8, P < 0.001]. When subjects
Discussion
The primary purpose of this study was to further examine the role of posterior precuneal and left prefrontal activation during episodic source memory retrieval using a similar source memory paradigm as in our previous study (Lundstrom et al., 2003) but with an increased latency between encoding and retrieval. Our major finding relates the levels of precuneal and left inferior prefrontal activations to the behavioral outcome of the source memory task. Specifically, activation of the left
Conclusion
These data provide support for the idea that the posterior precuneus is an integral part of successfully completed source memory retrieval and that both the precuneus and the left ventro-lateral frontal region/frontal operculum (BA 47) are crucial for correct source retrieval, especially when regeneration of rich episodic contextual associations is important. Further, data suggest three disparate regions in the left lateral prefrontal cortex, where the left ventro-lateral frontal region/frontal
Acknowledgments
Grants from the Swedish Medical Research Council VR-K2004-15074-01A, the Karolinska Institutet, the Swedish Medical Association, the Knut and Alice Wallenberg Foundation, and the Family Hedlund Foundation supported this work. B.N.L. was supported by 2 T32 GM 07266 Medical Scientist Training Program, National Institute of General Medical Sciences. We thank Guillen Fernandez for comments on an earlier version of the manuscript.
References (54)
- et al.
Functional–anatomic study of episodic retrieval using fMRI. I. Retrieval effort versus retrieval success
NeuroImage
(1998) - et al.
Executive control during episodic retrieval: multiple prefrontal processes subserve source memory
Neuron
(2002) - et al.
Memory orientation and success: separable neurocognitive components underlying episodic recognition
Neuropsychologia
(2003) - et al.
The mind's eye-precuneus activation in memory-related imagery
NeuroImage
(1995) - et al.
Generalisability, random effects, and population inference
NeuroImage
(1998) - et al.
Memory for perceived and imagined pictures—An event-related potential study
Neuropsychologia
(2002) - et al.
Neural correlates of episodic retrieval success
NeuroImage
(2000) - et al.
Isolating the retrieval of imagined pictures during episodic memory: activation of the left precuneus and left prefrontal cortex
NeuroImage
(2003) - et al.
The role of prefrontal cortex during tests of episodic memory
Trends Cogn. Sci.
(1998) - et al.
Functional specialization for semantic and phonological processing in the left inferior prefrontal cortex
NeuroImage
(1999)