Distinct roles for lateral and medial rostral prefrontal cortex in source monitoring of perceived and imagined events

Abstract

Rostral prefrontal cortex (PFC) is known to be involved in source memory, the ability to recollect contextual information about an event. However it is unclear whether subregions of rostral PFC may be differentially engaged during the recollection of different kinds of source detail. We used event related functional MRI to contrast two forms of source recollection: (1) recollection of whether stimuli had previously been perceived or imagined, and (2) recollection of which of two temporally distinct lists those stimuli had been presented in. Lateral regions of rostral PFC were activated in both tasks. However medial regions of rostral PFC were activated only when participants were required to recollect source information for self-generated, "imagined" stimuli, indicating a specific role in self-referential processing. In addition, reduced activity in a region of medial ventro-caudal PFC/basal forebrain was associated with making "imagined-to-perceived" confabulation errors. These results suggest that whilst the processing resources supported by some regions of lateral rostral PFC play a general role in source recollection, those supported by medial rostral PFC structures may be more specialised in their contributions.

Fig. 1

Examples of the stimuli used during study and test phases. In the study phase two temporally distinct lists of items were presented. In each list participants viewed either a clue and target word, or a clue and a question mark, prompting them to imagine the target word. In both cases they were instructed to count the number of letters in the target word. In the subsequent test phase they either viewed or were prompted to imagine target words embedded in sentences, and carried out one of two source tasks: either to recollect whether the word had been presented in the first or second list of the study phase (temporal source task), or whether the word had been seen or imagined in the study phase (perceived/imagined source task).

Fig. 2

Regions of rostral PFC activated in the critical contrasts. Activations in contiguous regions of bilateral rostral PFC were observed in comparisons of (A) Recollection of Perceived/Imagined and Temporal Source > New Items, and (B) Recollection of Temporal Source > Recollection of Perceived/Imagined Source. These regions are proposed to be involved in several different types of source recollection. By contrast medial regions of rostral PFC were only activated in comparisons of (C) Source Recollection for Items Imagined at Study > Source Recollection for Items Perceived at Study and (D) the interaction term [Imagined at Study (P/I–Temporal source)] − [Perceived at Study (P/I–Temporal source)]. These regions are assumed to be preferentially involved in source recollection involving self-generated information.

Fig. 3

Mean parameter estimates for medial BA 10 and medial ventro-caudal PFC for the temporal and P/I tasks according to study condition (perceived or imagined). Bars represent mean parameter estimates and error bars represent standard error of the mean.

Fig. 4

Scatter plot illustrating the correlation across participants between reduced activation in medial ventro-caudal PFC and likelihood of misattributing stimuli that were imagined at study and test as having been perceived.