Autobiographical memory is what people usually mean when they use the term "memory"; that is, the ability to remember past events from one's own life. Yet, the vast majority of functional neuroimaging studies of episodic memory have focused on memory for laboratory events, such as the appearances of words or pictures on a screen. Although laboratory paradigms can investigate most aspects of episodic memory in the scanner, autobiographical memory paradigms can expand on laboratory paradigms in several ways. First, autobiographical events typically involve a greater range in recollective qualities, such as self-reference, visuospatial imagery, and emotion. Second, autobiographical retrieval can be quite slow, thereby allowing the use of temporal differences to distinguish the contribution of various brain regions to retrieval processes. In collaboration with David Rubin and Kevin LaBar, we have been conducting fMRI studies of autobiographical memory that investigate recollective qualities and temporal dissociations. We have employed two different paradigms: (a) the photo paradigm, and (b) the cue-word paradigm.

• Cabeza, R., & St. Jacques, P. L. (2007). Functional neuroimaging of autobiographical memory. Trends in Cognitive Sciences, 11, 219-227.

The Photo Paradigm

In order to link the findings generated by fMRI studies of autobiographical memory to the large body of data generated by fMRI studies of episodic memory for laboratory events, it is critical to have control over several variables that tend to differ between autobiographical and laboratory paradigms, including strategic retrieval processes, the age of the memories, emotional content, and retrieval accuracy. To achieve this goal, we developed the photo paradigm, which allows the study of memories created in the real world (i.e., autobiographical memories) but under conditions in which the foregoing factors can be carefully controlled. In this paradigm, participants are provided with digital cameras and instructed to take photos in the real world (e.g., the university campus), and they are later scanned while remembering the photo-taking events using their own photos as cues. In one study (Cabeza et al., 2004), we found that, compared to recognizing photos taken by other participants, recognizing photos taken by oneself elicited greater activity in regions associated with self-referential processing (medial PFC), recollection (hippocampus), and visuospatial imagery (visual cortex, parahippocampal place area). In another study (St. Jacques et al., 2008), we took advantage of the photo paradigm to investigate one important form of relational memory, temporal-order memory, under conditions closer to memory in the real world. Whereas laboratory studies of temporal-order memory typically measure memory for the order of words in a list, which are impoverished stimuli separated by brief time intervals (e.g., seconds), temporal-order memory in the real world generally involves sensory-rich stimuli separated by longer intervals (e.g., minutes, hours). In our study, participants took pictures in a series of campus locations over a period of 5 hours, and in the scanner they were presented with two pictures and had to indicate which one they took first. To investigate the contribution of recollection vs. familiarity, we manipulated the lag between the two pictures, with short lags assumed to tap recollection and long lags assumed to tap familiarity. Consistent with other evidence, short lags (recollection) differentially recruited left PFC, hippocampal, and visual regions, whereas long lags (familiarity) differentially recruited right PFC.

• St. Jacques, P. L., Rubin, D. C., LaBar, K. S., & Cabeza, R. (2008). The short and long of it: Neural correlates of temporal-order memory for autobiographical events. Journal of Cognitive Neuroscience, 20, 1327-1341.
• Cabeza, R., Prince, S. E., Daselaar, S. M., Greenberg, D., Budde, M., Dolcos, F., LaBar, K. S., & Rubin, D. C. (2004). Brain activity during episodic retrieval of autobiographical and laboratory events: An fMRI study using a novel photo paradigm. Journal of Cognitive Neuroscience, 9, 1533-1594.
  

The Cue-word Paradigm

In another group of autobiographical fMRI studies, we used the Crovitz-Galton technique, in which participants generate memories in response to simple word cues. Because the generation process is slow, this paradigm allows the use differences in the temporal pattern of fMRI activity to associate brain regions with different retrieval processes. In one study, participants pressed a response key when they accessed an autobiographical memory (mean: 12.5 sec), then continued elaborating the memory until they were presented with emotion and vividness scales. Whereas activity in the hippocampus and right PFC regions was greater before than after the key press, indicating the involvement of these regions in memory access processes, activity in visual cortex and left PFC regions were greater after the key press, suggesting the contribution of these areas to memory elaboration processes. Using the rating scales, we were also able to identify regions involved in processing the emotional content of autobiographical memories, such as the amygdala, and regions involved in processing the vividness of these memories, such as visual cortex (Daselaar et al., 2008). We plan to follow up these studies in older adults.

• Daselaar, S. M., Rice, H. J., Greenberg, D. L., Cabeza, R., LaBar, K. S., & Rubin, D. C. (2008). The spatiotemporal dynamics of autobiographical memory: Neural correlates of recall, emotional intensity, and reliving. Cerebral Cortex, 18, 217-229.