Orbitofrontal and hippocampal contributions to memory for face-name associations: the rewarding power of a smile

Abstract

Memory processes can be enhanced by reward, and social signals such a smiling face can be rewarding to humans. Using event-related functional MRI (fMRI), we investigated the rewarding effect of a simple smile during the encoding and retrieval of face-name associations. During encoding, participants viewed smiling or neutral faces, each paired with a name, and during retrieval, only names were presented, and participants retrieved the associated facial expressions. Successful memory activity of face-name associations was identified by comparing remembered vs. forgotten trials during both encoding and retrieval, and the effect of a smile was identified by comparing successful memory trials for smiling vs. neutral faces. The study yielded three main findings. First, behavioral results showed that the retrieval of face-name associations was more accurate and faster for smiling than neutral faces. Second, the orbitofrontal cortex and the hippocampus showed successful encoding and retrieval activations, which were greater for smiling than neutral faces. Third, functional connectivity between the orbitofrontal cortex and the hippocampus during successful encoding and retrieval was stronger for smiling than neutral faces. As a part of the reward system, the orbitofrontal cortex may modulate memory processes of face-name associations mediated by the hippocampus. Interestingly, the effect of a smile during retrieval was found even though only names were presented as retrieval cues, suggesting that the effect was mediated by face imagery. Taken together, the results demonstrate how rewarding social signals from a smiling face can enhance relational memory for face-name associations.

Fig. 1

Task paradigm. During encoding, subjects were required to learn face-name associations, and to rate facial expressions with four response options (DH: definitely happy, PH: probably happy, PN: probably neutral, DN: definitely neutral). During retrieval, subjects were presented with studied and new names, and were required to retrieve the facial expressions associated with the names. Subjects responded with four response options (Hap: happy face, Neu: neutral face, UK: studied face of unknown expression, New: not studied face (i.e., New Name).

Fig. 2

Behavioral data at the retrieval phase. Error bars represent standard error. (A) Hit rate [Hit/(Hit+Miss)]. (B) Reaction time (ms). Hit: successful retrieval of facial expressions and names, Miss: missed retrieval of facial expressions and successful retrieval of names, **: p<0.01, *: p<0.05.

Fig. 3

Activation images and effect sizes of right orbitofrontal cortex (OFC) and hippocampus (HIP) in the successful encoding of face-name associations for happy relative to neutral expressions. Error bars represent standard error. (A) Activation and effect sizes in the right orbitofrontal cortex. (B) Activation and effect sizes in the right hippocampus. *: p<0.05.

Fig. 4

Activation images and effect sizes of bilateral orbitofrontal cortices (OFC), right hippocampus (HIP) and left parahippocampal gyrus (PHG) in the successful retrieval of face-name associations for happy relative to neutral expressions. Error bars represent standard error. (A) Activation and effect sizes in the bilateral orbitofrontal cortices. (B) Activation and effect sizes in the right hippocampus and left parahippocampal gyrus. **: p<0.01, *: p<0.05.