Electrophysiological correlates of processes supporting memory for faces

Abstract

The retrieval processes supporting recognition memory for faces were investigated using event-related potentials (ERPs). The focus for analyses was ERP old/new effects, which are the differences between neural activities associated with correct judgments to old (studied) and new (unstudied) test stimuli. In two experiments it was possible to identify three old/new effects that behaved as neural indices of the process of recollection. In both experiments there was one old/new effect that behaved as an index of the process of familiarity. These outcomes are relevant to the ongoing debate about the functional significance of ERP old/new effects and the implications that scalp-recorded electrophysiological data have for theories of the processes supporting long-term memory judgments.

Target-context unitization effect on the familiarity-related FN400: A face recognition exclusion task

“…3: background-exclusion task). These figures show two main peaks similar to those previously reported for recognition tasks using non-famous faces (e.g., Guillaume et al, 2012a,b;Guillaume and Tiberghien, 2013;MacKenzie and Donaldson, 2009;Yick and Wilding, 2014;Yovel and Paller, 2004). They correspond to the ERP recognition effects usually described in the literature as FN400 or mid-frontal old-new effect (300-500 ms) and the late positive component (LPC) corresponding to the parietal old-new effect (500-800 ms).…”

“…In a similar design, MacKenzie and Donaldson (2007); MacKenzie and Donaldson (2009) obtained yet a different set of electrophysiological outcomes, finding that the mid-frontal old-new effect (300-500 ms) reflected recollection of contextual details (occupation or other details), whereas the posterior effect, in the same epoch, was statistically equivalent whether or not contextual details were recovered. More recently, Yick and Wilding (2014) presented findings comparable to those of MacKenzie and Donaldson (2007) with a manipulation of face-intrinsic features (e.g., a white bar superimposed on the face: eyes vs. mouth). In the early epoch (300-500 ms), they reported that frontal modulation was larger when accompanied by memory for context, whereas a posterior distributed modulation was of equal size for all recognized faces.…”

“…The 300-500 ms and 500-800 ms epochs were selected a priori on the basis of approaches and findings in previous ERP studies of face recognition. Analyses were conducted in time windows and at electrode sites where it has been proposed that ERPs index the processes of familiarity and recollection (Guillaume and Tiberghien, 2013;MacKenzie and Donaldson, 2007;Yick and Wilding, 2014). For familiarity, the focused analyses were conducted at three mid-frontal electrodes (F3, Fz, F4) in the epoch between 300 and 500 ms based on the suggestion that early frontal effects might index familiarity.…”

“…For familiarity, the focused analyses were conducted at three mid-frontal electrodes (F3, Fz, F4) in the epoch between 300 and 500 ms based on the suggestion that early frontal effects might index familiarity. For recollection, the analyses were performed on data from three parietal electrodes (P3, Pz, P4) as well as the three mid-frontal electrodes, as a frontal modulation has been observed with faces as test stimuli in the epoch between 500 and 800 ms (MacKenzie and Donaldson, 2007;Yick and Wilding, 2014).…”

“…To our knowledge, the influence of the nature of the perceptual information associated to the target has not been directly investigated in face recognition. In the studies described above, participants were asked to make an old-new judgment and, for recognized faces, to report contextual details about the study episode: name, occupation, or the location of a specific feature associated with the face at study (MacKenzie and Donaldson, 2007;MacKenzie and Donaldson, 2009;Yick and Wilding, 2014;Yovel and Paller, 2004). Importantly, in these designs, the specific information that was associated to the studied face was presented at study but absent at test.…”