Research report
Distinct prefrontal cortex activity associated with item memory and source memory for visual shapes

https://doi.org/10.1016/S0926-6410(03)00082-XGet rights and content

Abstract

In contrast to item memory, which refers to recognition or recall of previously presented information, source memory refers to memory for the context of previously presented information. The relatively few functional MRI (fMRI) source memory studies conducted to date have provided evidence that item memory and source memory are associated with differential activity in right and left prefrontal cortex, respectively. To both confirm this distinction in prefrontal cortex and to determine whether other differences in the neural substrates associated with these cognitive functions exist, an event-related fMRI study was conducted. In this study, item memory and source memory encoding phases were identical; participants viewed a series of abstract visual shapes presented on the left or right side of the screen and were instructed to remember each shape and its spatial location. During the item memory retrieval phase, shapes from the encoding phase were intermixed with new shapes and participants made an old–new judgment. During the source memory retrieval phase, all shapes were from the encoding phase and participants made a left–right judgment. An event-related analysis of item memory and source memory revealed a right and left prefrontal cortex distinction. Moreover, only item memory was associated with activity in the medial temporal lobes. These results confirm and extend previous findings that item memory and source memory are associated with distinct neural substrates.

Introduction

One can determine whether a person remembers a previously presented item by showing them the item and asking “Have you seen this before?” Alternatively, one can inquire as to the context in which the item was previously presented, e.g. “Where did you see this before?” Here, the term item memory will be used to refer to memory of previous exposure to a specific item1 while the term source memory will be used to refer to the context in which that item was presented [20], [49]. Source memory can refer to the time at which information was presented (e.g. words from list 1 vs. list 2), a feature of the information (e.g. red vs. green shapes), or the spatial location of previously presented information (e.g. objects on the left vs. right side of the screen). In the last decade, a growing body of convergent evidence has been amassed regarding the neural substrates associated with item memory; however, much less is known about the neural substrates associated with source memory.

Using functional magnetic resonance imaging (fMRI) and positron emission tomography (PET), item memory has been shown to consistently activate right prefrontal cortex [2], [3], [4], [14], [15], [16], [26], [27], [32], [33], [38], [51], [57], consistent with Tulving’s hemispheric encoding/retrieval asymmetry model (HERA) [31], [55], the medial parietal cortex [2], [3], [5], [14], [15], [16], [19], [21], [27], [41], [45], [51], [56], and the medial temporal lobes, which include the hippocampus and parahippocampal gyri [4], [5], [6], [9], [13], [24], [37], [39], [40], [41], [42], [43], [44], [45], [46], [51].

Findings from event-related potential (ERP) studies of item memory are highly consistent with the functional imaging results. In particular, item memory elicits a right frontal [1], [7], [8], [34], [53], [58], [59] and parietal scalp response [1], [7], [8], [34], [58], [59]. To date, medial temporal lobe activation during item memory has not been reported using ERPs, presumably due to the depth of medial temporal sources relative to recording electrodes on the scalp.

In contrast to item memory, the source memory ERP literature is less consistent. Some researchers have argued that source memory and item memory have similar scalp distributions, differing only quantitatively, rather than qualitatively, and thus rely on the same neural substrates [1], [7], [59]. In these studies, participants were instructed to make an item memory judgment immediately followed by a source memory judgment for each stimulus thus requiring a dual judgment; the concomitant item memory and source memory retrieval processes on each trial may have masked a true distinction between item memory and source memory. Indeed, neuropsychological evidence of source memory deficits with spared item memory [18], [48] indicates there are distinct neural substrates for item memory and source memory. In ERP studies where item and source memory decisions have been isolated, source memory has been associated with unique prefrontal activation [47], [53], specifically in the left prefrontal cortex [34]. Although ERPs provide excellent temporal resolution, their spatial resolution is poor relative to fMRI (i.e. centimeters vs. millimeters); thus, fMRI results should provide a more spatially precise measure of neural activity associated with item memory and source memory.

Though few in number, fMRI studies of item memory and source memory have produced very consistent results. Two studies used words [32], [38] and one study used objects [33]. Across studies, item memory was associated with activity in the right prefrontal cortex and source memory was associated with activity in the left prefrontal cortex. Of note, fMRI studies of either item memory or source memory have not generally isolated the processes associated with successful memory retrieval from those associated with memory retrieval effort. In the present study, this issue was addressed by only reporting results of comparisons between correct responses (e.g. correct item memory trials compared to correct rejection trials). This type of analysis has been conducted in a remember/know paradigm [17], where remember judgments, compared to know judgments, were associated with left prefrontal activity. The similarity between these results and source memory results might be expected, as remember judgments have been associated with memory for specific contextual details [54].

The purpose of the present investigation was to better delineate the neural distinction between item memory and source memory. To more purely assess aspects of memory, without the neural activity associated with the processing of meaningful stimuli (e.g. words or objects), abstract visual shapes were used. We found distinct neural substrates associated with item memory and source memory in both the prefrontal cortex and the medial temporal lobe.

Section snippets

Participants

Eight right-handed adults (six females), ranging in age from 25 to 45 years, took part in the experiment. All participants had normal or corrected to normal visual acuity. After the nature of the experiment, which had been approved by the Johns Hopkins internal review board, was fully comprehended by each participant, written consent was obtained before the experiment commenced.

Stimuli and tasks

Abstract visual shapes, similar to those used to study visuospatial processing [22], [23], [50], were selected as

Results

Behavioral accuracies for item memory hits (79±6%, one standard error reported), correct rejections (79±6%), and source memory hits (77±3%) indicate that the task was not too easy, nor too difficult. In addition, the similarity in accuracy across trial types argues against difficulty related differences that could potentially contaminate the results.

Fig. 2 illustrates the activity associated with item memory and source memory. Consistent with a large body of experimental evidence, item memory

Discussion

In the present study, item memory, as assessed by the item memory vs. correct rejection contrast, was associated with activity in right prefrontal cortex, parietal cortex, and the medial temporal lobe consistent with a large body of neuroimaging research [2], [3], [4], [5], [6], [9], [13], [14], [15], [16], [19], [21], [24], [26], [27], [32], [33], [37], [39], [40], [41], [42], [43], [44], [45], [46], [51], [55], [56], [57]. Source memory, as assessed by the source memory vs. item memory

Acknowledgments

We would like to thank Terri Brawner for her expert assistance with MRI scanning.

References (59)

  • M. Lotze et al.

    fMRI evaluation of somatotopic representation in human primary motor cortex

    Neuroimage

    (2000)
  • C. Ranganath et al.

    Frontal brain potentials during recognition are modulated by requirements to retrieve perceptual detail

    Neuron

    (1999)
  • S.A.R.B. Rombouts et al.

    Anterior medial temporal lobe activation during attempted retrieval of encoded visuospatial sciences: An event-related fMRI study

    Neuroimage

    (2001)
  • M.D. Rugg et al.

    The role of the prefrontal cortex in recognition memory and memory for source: A fMRI study

    Neuroimage

    (1999)
  • D.L. Schacter et al.

    Late onset of anterior prefrontal activity during true and false recognition: An event-related fMRI study

    Neuroimage

    (1997)
  • D.L. Schacter et al.

    Neuroanatomical correlates of veridical and illusory recognition memory: Evidence from positron emission tomography

    Neuron

    (1996)
  • D.J. Wegesin et al.

    Age-related changes in source memory retrieval: an ERP replication and extension

    Cogn. Brain. Res.

    (2002)
  • K. Allen et al.

    Neural correlates of cued recall with and without retrieval of source memory

    Neuroreport

    (1998)
  • N.C. Andreasen et al.

    Short-term and long-term verbal memory: A positron emission tomography study

    Proc. Natl. Acad. Sci. USA

    (1995)
  • R.L. Buckner et al.

    Functional anatomical studies of explicit and implicit memory retrieval tasks

    J. Neurosci.

    (1995)
  • R. Cabeza et al.

    Can medial temporal lobe regions distinguish true from false? An event-related functional MRI study of veridical and illusory recognition memory

    Proc. Natl. Acad. Sci. USA

    (2001)
  • L.L. Eldridge et al.

    Remembering episodes: A selective role for the hippocampus during retrieval

    Nat. Neurosci.

    (2000)
  • J.D.E. Gabrieli et al.

    Separate neural bases of two fundamental memory processes in the human medial temporal lobe

    Science

    (1997)
  • C.L. Grady et al.

    Age-related reductions in human recognition memory due to impaired encoding

    Science

    (1995)
  • J.V. Haxby et al.

    Face encoding and recognition in the human brain

    Proc. Natl. Acad. Sci. USA

    (1996)
  • R.N.A. Henson et al.

    Confidence in recognition memory for words: Dissociating right prefrontal roles in episodic retrieval

    J. Cogn. Neurosci.

    (2000)
  • R.N.A. Henson et al.

    Recollection and familiarity in recognition memory: An event-related functional magnetic resonance imaging study

    J. Neurosci.

    (1999)
  • M.K. Johnson et al.

    Source monitoring

    Psychol. Bull.

    (1993)
  • S.M. Kosslyn

    Seeing and imagining in the cerebral hemispheres: A computational approach

    Psychol. Rev.

    (1987)
  • Cited by (0)

    View full text