In search of recollection and familiarity signals in the hippocampus

Abstract

fMRI studies of recognition memory have often been interpreted to mean that the hippocampus selectively subserves recollection and that adjacent regions selectively subserve familiarity. Yet, many of these studies have confounded recollection and familiarity with strong and weak memories. In a source memory experiment, we compared correct source judgments (which reflect recollection) and incorrect source judgments (often thought to reflect familiarity) while equating for old-new memory strength by including only high-confidence hits in the analysis. Hippocampal activity associated with both correct source judgments and incorrect source judgments exceeded the activity associated with forgotten items and did so to a similar extent. Further, hippocampal activity was greater for high-confidence old decisions relative to forgotten items even when source decisions were at chance. These results identify a recollection signal in the hippocampus and may identify a familiarity signal as well. Similar results were obtained in the parahippocampal gyrus. Unlike in the medial temporal lobe, activation in prefrontal cortex increased differentially in association with source recollection.

Figure 1

Example of test procedure and behavioral analysis: The stimulus “wheel” was studied under the common contextual-cue condition and endorsed as a high-confidence hit (5 or 6) in the old–new recognition task. For the subsequent source memory task, a response of 1 or 2 indicates false source memory, a response of 3 or 4 indicates source guesses, and a response of 5 or 6 indicates true source memory (source responses 1, 2, and 3 are incorrect and source responses 4, 5, and 6 are correct).

Figure 2

Proportion of responses to targets and foils for each confidence level in the old–new task, n = 16, mean recognition d′ = 1.93 (0.15).

Figure 3

Activation identified for separate contrasts of correct source judgments versus forgotten items and incorrect source judgments versus forgotten items. To equate for memory strength, the source-correct and source-incorrect data were based on old decisions made with high confidence. Error bars for the two source categories represent the SEM of the difference scores for each comparison, whereas the error bar for the forgotten items represents the root mean square of the SEM values associated with the two individual comparisons (* denotes a significant difference relative to forgotten items, p-corrected < .05).

Figure 4

(A) Activation identified in the hippocampus for the contrast of true source judgments versus forgotten items. (B) Activation identified in the hippocampus for the contrast of source guesses versus forgotten items. Bar graphs show signal correlated with forgotten items and with recognition decisions that were accompanied by true, guess, or false source judgments. Also illustrated are the time courses of the hrf correlated with these results and MRI sections locating the activation on averaged anatomical images for our participants. Error bars for the three source categories represent the SEM of the difference scores for each comparison (relative to forgotten items), whereas the error bar for the forgotten items represents the root mean square of the SEM values associated with the three individual comparisons (* denotes a significant difference relative to forgotten items, p-corrected < .05).

Figure 4

(A) Activation identified in the hippocampus for the contrast of true source judgments versus forgotten items. (B) Activation identified in the hippocampus for the contrast of source guesses versus forgotten items. Bar graphs show signal correlated with forgotten items and with recognition decisions that were accompanied by true, guess, or false source judgments. Also illustrated are the time courses of the hrf correlated with these results and MRI sections locating the activation on averaged anatomical images for our participants. Error bars for the three source categories represent the SEM of the difference scores for each comparison (relative to forgotten items), whereas the error bar for the forgotten items represents the root mean square of the SEM values associated with the three individual comparisons (* denotes a significant difference relative to forgotten items, p-corrected < .05).

Figure 5

(A) Activation identified in left parahippocampal cortex (sagittal image) for the contrast of true source decisions versus forgotten items. (B) Activation identified in left perirhinal cortex (coronal image) for the contrast of source guesses versus forgotten items. Bar graphs show signal correlated with forgotten items and with recognition decisions that were accompanied by true, guess, or false source judgments. Also illustrated are the time courses of the hrf correlated with these results, and coronal sections locating the activation on averaged anatomical images for our participants. Error bars for the three source categories represent the SEM of the difference scores for each comparison (relative to forgotten items), whereas the error bar for the forgotten items represents the root mean square of the SEM values associated with the three individual comparisons (* denotes a significant difference relative to forgotten items, p-corrected < .05).

Figure 5

(A) Activation identified in left parahippocampal cortex (sagittal image) for the contrast of true source decisions versus forgotten items. (B) Activation identified in left perirhinal cortex (coronal image) for the contrast of source guesses versus forgotten items. Bar graphs show signal correlated with forgotten items and with recognition decisions that were accompanied by true, guess, or false source judgments. Also illustrated are the time courses of the hrf correlated with these results, and coronal sections locating the activation on averaged anatomical images for our participants. Error bars for the three source categories represent the SEM of the difference scores for each comparison (relative to forgotten items), whereas the error bar for the forgotten items represents the root mean square of the SEM values associated with the three individual comparisons (* denotes a significant difference relative to forgotten items, p-corrected < .05).

Figure 6

Activation identified in prefrontal cortex for the contrast of high-confidence hits that were accompanied by true source judgments versus high-confidence hits that were accompanied by source guesses. (A) Left VLPFC (BA 45) and (B) right DLPFC (BA 46). Error bars represent the SEM of the difference scores (* denotes a significant difference between true source and source guesses, p-corrected < .05).

Figure 7

Activation identified for the contrast of hits rated as old–new “6” versus forgotten items. Error bars represent the SEM of the difference scores (* denotes a significant difference relative to forgotten items, p-corrected < .05).