Elsevier

Neuropsychologia

Volume 56, April 2014, Pages 26-36
Neuropsychologia

Effects of attractiveness on face memory separated from distinctiveness: Evidence from event-related brain potentials

https://doi.org/10.1016/j.neuropsychologia.2013.12.023Get rights and content

Highlights

  • We tested attractiveness effects on face memory irrespective of distinctiveness.

  • We found more accurate memory for unattractive than attractive faces.

  • Attractive faces elicited a larger N250/EPN ERP component.

  • This ERP effect during learning correlated with the memory effect at test.

  • Affective processing of attractive faces seems to hamper encoding into memory.

Abstract

The present study examined effects of attractiveness on behavioral and event-related potential (ERP) correlates of face memory. Extending previous reports, we controlled for potential moderating effects of distinctiveness, a variable known to affect memory. Attractive and unattractive faces were selected on the basis of a rating study, and were matched for distinctiveness. In a subsequent recognition memory experiment, we found more accurate memory for unattractive relative to attractive faces. Additionally, an attractiveness effect in the early posterior negativity (EPN) during learning, with larger amplitudes for attractive than unattractive faces, correlated significantly with the magnitude of the memory advantage for unattractive faces at test. These findings establish a contribution of attractiveness to face memory over and above the well-known effect of distinctiveness. Additionally, as the EPN is typically enhanced for affective stimuli, our ERP results imply that the processing of emotionally relevant attractive faces during learning may hamper their encoding into memory.

Introduction

Attractive people profit from multiple advantages in social interactions due to their appealing looks. For instance, attractive children achieve both better marks and more attention in class (Lerner & Lerner, 1977), and as adults, beautiful individuals are more successful in their professions and are helped more readily in dire situations (Benson et al., 1976, Harrell, 1978, Mims et al., 1975). Whereas attractiveness is thus usually considered a highly desirable attribute, it is largely unclear whether attractive people are also particularly memorable.

A number of studies examined memory for attractive and unattractive faces, with rather discrepant results. Several studies reported more accurate memory for attractive relative to unattractive faces (Cross et al., 1971, Marzi and Viggiano, 2010, Zhang et al., 2011), whereas others either found the opposite pattern (Light et al., 1981, Sarno and Alley, 1997), or no difference in memory (Brigham, 1990, Wickham and Morris, 2003). We considered that this inconsistency may be partly related to other facial characteristics which were not systematically controlled in previous research. A particularly important characteristic in this context is distinctiveness. Highly distinctive faces strongly deviate from an average or prototypical face, as they, for instance, contain unusually sized or shaped facial features (such as particularly small eyes), or unusual facial texture or coloration. By contrast, less distinctive or typical faces are perceptually closer to the prototype. Importantly, it is well known that distinctive faces are remembered particularly well (e.g., Valentine, 1991), such that more accurate memory for either attractive or unattractive faces in previous studies may have been influenced by differences in distinctiveness in the respective sets of faces.

Most theoretical accounts on facial attractiveness suggest a systematic relationship to distinctiveness. The direction of this relation, however, is a matter of scientific debate. Initially, it had been suggested that attractive faces are ‘average’, and thus typical rather than distinctive. This suggestion is supported by results indicating that an average face, created by merging a number of individual faces into a single face morph, is typically rated as more attractive than the individual faces that constitute the morph (Langlois & Roggman, 1990). Averageness per se, however, cannot fully explain why certain faces are judged as more attractive than others. For instance, DeBruine, Jones, Unger, Little, and Feinberg (2007) demonstrated that a morph across highly attractive faces was rated as more attractive than a morph across randomly chosen faces. In addition, systematic deviations from the average towards increased sexual dimorphisms (such as pronounced cheek bones in male faces) are perceived as more attractive than average characteristics (Perrett et al., 1998). Recently, it has been reported that averageness is attractive in some dimensions (especially those related to shape information) but not in others (particularly those related to reflectance information; Said & Todorov, 2011). In sum, it seems justified to conclude that distinctiveness (or averageness) explains variance in attractiveness ratings, and thus may affect memory differences between attractive and unattractive faces. Critically, the main question for the present study was whether and how attractiveness modulates face memory, even when distinctiveness is held constant for attractive and unattractive faces.

Distinctiveness is traditionally measured with so-called face-in-the-crowd (FITC) ratings, in which participants indicate how likely they would spot a given face in a crowd of people (e.g., Valentine & Endo, 1992). This measure, however, may not be ideal in the present context, as attractiveness potentially biases such ratings, i.e., participants may have a tendency to indicate a high probability of detecting a face in a crowd because it is attractive, independent of whether the face is typical or distinct (‘I would notice such a beautiful person’). Deviation measures have therefore been introduced as an alternative to assess distinctiveness (Wickham & Morris, 2003). Here, participants rate how strongly a face deviates from typical or average familiar faces. We considered that these two measures of distinctiveness might only partially relate to the same construct, and would not necessarily relate to attractiveness in the same manner.

Apart from distinctiveness, other factors may also influence memory for attractive versus unattractive faces. For instance, a recent socio-cognitive account suggests that participants’ motivation to individuate faces affects memory (Hugenberg, Young, Bernstein, & Sacco, 2010). In support of this, participants have been reported to more accurately recognize faces randomly assigned to their own university, compared to faces assigned to a different university (Bernstein, Young, & Hugenberg, 2007), a finding which was thought to reflect the higher motivation to individualize social in- relative to out-group faces. Of relevance, it has been reported that attractive faces are more likely categorized as belonging to the viewer′s social in- rather than out-group (Johnson, 1981), and that participants are more motivated to view attractive faces (Aharon et al., 2001). Accordingly, socio-cognitive accounts would predict more accurate memory for attractive relative to unattractive faces, an effect which should be more likely detected if other contributing factors such as distinctiveness are controlled for.

Importantly, behavioral measures of recognition memory only depict the outcome of a cascade of cognitive sub-processes. By contrast, the analysis of event-related potentials (ERPs) allows investigating neural correlates of these processes as they unfold over time, and by inference to identify the underlying cognitive or affective mechanisms. The first component typically examined in face processing studies is the P1, a positive peak at occipital sites occurring approximately 100 ms post stimulus presentation. This component is sensitive to physical stimulus characteristics, such as luminance or contrast (Luck, 2005), and may thus reflect early stages of visual processing.

The following occipito-temporal N170 (Bentin, Allison, Puce, Perez, & McCarthy, 1996), a negative peak at approximately 170 ms, is typically larger for face stimuli relative to other object classes (see e.g., Eimer, 2011). It is commonly assumed to reflect structural face encoding (Eimer, 2000) or the detection of a face-like pattern in the visual field (Amihai et al., 2011, Schweinberger and Burton, 2003). Previous reports of attractiveness effects on the N170 are inconsistent, with some studies reporting larger N170 to unattractive faces (experiment 2 in Halit, de Haan, & Johnson, 2000), others reporting larger amplitudes to attractive faces (Marzi & Viggiano, 2010), and still others reporting no difference (Roye et al., 2008, Schacht et al., 2008).

Subsequent to N170, a positive occipito-temporal peak is observed at approximately 220–250 ms. This P2 has been interpreted to reflect the processing of metric distances between facial features (e.g., Latinus & Taylor, 2006) or the perceived typicality of a face (e.g., Wiese, Kaufmann, & Schweinberger, in press), and is modulated by an observer′s expertise with a specific facial category (Stahl, Wiese, & Schweinberger, 2008). Interestingly, a number of recent studies found smaller P2 components for naturally distinctive relative to more typical faces (Schulz, Kaufmann, Kurt, & Schweinberger, 2012), as well as for spatial caricatures relative to veridical faces (Kaufmann and Schweinberger, 2012, Schulz et al., 2012). To the best of our knowledge, no previous study examined attractiveness effects on P2 using natural faces (though Halit et al. (2000, experiment 1), report a smaller P2 for unattractive faces with artificially increased distances between features).

The following negative wave, the N250, is known to be larger for learned relative to novel facial identities (Kaufmann et al., 2009, Tanaka et al., 2006). Previous recognition memory studies observed stronger memory effects in the N250 time range for specific categories of faces, for instance for young relative to old faces in young participants (e.g., Wiese, Schweinberger, & Hansen, 2008). Moreover, distinctive faces have been reported to elicit larger N250 amplitudes than typical faces (Schulz et al., 2012, Schulz et al., 2012). Importantly, previous studies also reported more negative amplitudes for attractive relative to unattractive faces in the so-called early posterior negativity (EPN; e.g., Werheid, Schacht, & Sommer, 2007). EPN is typically interpreted as reflecting enhanced perceptual processing of affective stimuli, and has been found to be larger for emotional relative to neutral faces (Rellecke et al., 2012, Schupp et al., 2004) and for affective relative to neutral pictures (Schupp et al., 2007). The N250 and EPN originate from different research traditions, and may therefore be assumed to reflect different processes. However, they substantially overlap with respect to timing and scalp topography, which may complicate an unequivocal interpretation of a given ERP effect as reflecting an N250 or EPN in this time range. For the present study, it was of particular interest whether or not any larger N250/EPN amplitudes for attractive faces in previous studies were related to enhanced distinctiveness.

Finally, a late posterior component (LPC) typically occurs at centro-parietal scalp sites and is maximal between 300 and 700 ms. The LPC has been observed to be larger for affective than neutral pictures (Schupp et al., 2007), for emotional than neutral facial expressions (Schacht et al., 2008, Schupp et al., 2004), and for distinctive than typical faces (Schulz et al., 2012, Schulz et al., 2012). The LPC has been interpreted as reflecting an engagement towards motivationally significant stimuli (Cuthbert, Schupp, Bradley, Birbaumer, & Lang, 2000), and has also been shown to be larger for attractive relative to unattractive faces (Schacht et al., 2008, Werheid et al., 2007). Importantly, ERP effects of episodic memory occur in a similar time range. These so-called old/new effects reflect more positive amplitudes for correctly remembered learned (i.e., hits) relative to correctly rejected new items (i.e., correct rejections) and are commonly subdivided into an earlier frontal (300–500 ms) and a later more posterior effect (500–700 ms; see e.g., Rugg & Curran, 2007). While most researchers agree that the later old/new effect reflects recollection, i.e., the conscious retrieval of study phase detail, the exact processes reflected by the earlier effect are still debated (e.g., Curran, 2000, Paller et al., 2007). Of particular interest, Marzi and Viggiano (2010) found an early old/new effect for attractive but not for unattractive faces, which they interpreted as enhanced familiarity-based recognition, i.e., a stronger feeling of knowing for attractive faces.

In light of the literature discussed above, the aims of the present study were threefold: First, we aimed at determining effects of attractiveness on face memory while controlling for effects of distinctiveness. In doing so, we reasoned that the discrepancies regarding effects of attractiveness on face memory may well reflect differential confounds with distinctiveness that had been largely neglected in earlier research. Second, we aimed to test whether attractive relative to unattractive faces elicited larger amplitudes in the N250/EPN and LPC components when stimuli are matched for distinctiveness, as previous results could similarly have been confounded by this variable. Finally, the only previous study on ERP correlates of memory for attractive versus unattractive faces revealed an early old/new effect for attractive faces only. Again, we wanted to examine whether this effect was related to distinctiveness. To address these issues, we first conducted a rating experiment in which two measures of distinctiveness (deviation and FITC) and attractiveness were assessed in a large set of face stimuli. In a second step, we then tested ERP correlates of recognition memory for subsets of attractive and unattractive faces that were matched for distinctiveness.

Section snippets

Participants

Twenty participants (18–35 years; M=23.8 years±3.08SD; 10 females) were recruited and gave written informed consent prior to testing. All participants reported normal or corrected to normal vision and were right-handed as assessed by the Edinburgh Handedness Inventory (Oldfield, 1971). None of the participants reported neurological or psychiatric disorders. Participants received either course credit or monetary reimbursement after testing.

Stimuli

We collected 560 full-color facial photographs of young

Participants

We recruited 20 participants (10 females, mean age=24 years±2.4SD, all right-handed according to the Edinburgh Handedness Inventory), who had not participated in the rating experiment. Data from five additional participants had to be excluded from analyses due to insufficient trial numbers for ERP averaging (N<15) and/or chance performance in the memory task. All participants reported normal or corrected to normal vision, and none reported neurological or psychiatric disorders. All participants

Discussion

The present study examined effects of attractiveness on face memory. For that purpose we assessed two measures of distinctiveness (FITC and deviation) as well as attractiveness for a large set of faces in a rating study, and used a subset of these stimuli that were matched for deviation scores in a recognition paradigm. With respect to the first aim stated in Section 1, we observed an effect of attractiveness on face memory (as measured with d′), reflecting more accurate recognition of

Acknowledgment

This study was supported by a grant of the Deutsche Forschungsgemeinschaft (DFG) to H.W. and S.R.S. (WI 3219/5-2). We are thankful for help during data acquisition by Kathrin Rauscher.

References (63)

  • J. Rellecke et al.

    Does processing of emotional facial expressions depend on intention? Time-resolved evidence from event-related brain potentials

    Biological Psychology

    (2012)
  • B. Rossion et al.

    Does physical interstimulus variance account for early electrophysiological face sensitive responses in the human brain? Ten lessons on the N170

    Neuroimage

    (2008)
  • M.D. Rugg et al.

    Event-related potentials and recognition memory

    Trends in Cognitive Sciences

    (2007)
  • C. Schulz et al.

    Faces forming traces: Neurophysiological correlates of learning naturally distinctive and caricatured faces

    Neuroimage

    (2012)
  • C. Schulz et al.

    Effects of anticaricaturing vs. caricaturing and their neural correlates elucidate a role of shape for face learning

    Neuropsychologia

    (2012)
  • S.R. Schweinberger et al.

    Covert recognition and the neural system for face processing

    Cortex

    (2003)
  • K. Werheid et al.

    Facial attractiveness modulates early and late event-related brain potentials

    Biological Psychology

    (2007)
  • H. Wiese et al.

    The age of the beholder: ERP evidence of an own-age bias in face memory

    Neuropsychologia

    (2008)
  • I. Amihai et al.

    Neural adaptation is related to face repetition irrespective of identity: A reappraisal of the N170 effect

    Experimental Brain Research

    (2011)
  • J.Y. Baudouin et al.

    When the smile is a cue to familiarity

    Memory

    (2000)
  • S. Bentin et al.

    Electrophysiological studies of face perception in humans

    Journal of Cognitive Neuroscience

    (1996)
  • M.J. Bernstein et al.

    The cross-category effect: Mere social categorization is sufficient to elicit an own-group bias in face recognition

    Psychological Science

    (2007)
  • J.C. Brigham

    Target person distinctiveness and attractiveness as moderator variables in the confidence accuracy relationship in eyewitness identifications

    Basic and Applied Social Psychology

    (1990)
  • P.M. Bronstad et al.

    Beautry is in the 'we' of the beholder: greater agreement on facial attractiveness among close relations

    Perception

    (2007)
  • J.F. Cross et al.

    Sex, race, age, and beauty as factors in recognition of faces

    Perception & Psychophysics

    (1971)

    (6)

    (1971)
  • T. Curran

    Brain potentials of recollection and familiarity

    Memory & Cognition

    (2000)
  • L.M. DeBruine et al.

    Dissociating averageness and attractiveness: attractive faces are not always average

    Journal of Experimental Psychology: Human Perception and Performance

    (2007)
  • M. Eimer

    The face-specific N170 component reflects late stages in the structural encoding of faces

    Neuroreport

    (2000)
  • M. Eimer

    The face-sensitive N170 component of the event-related brain potential

  • J.M. Gardiner et al.

    How level of processing really influences awareness in recognition memory

    Canadian Journal of Experimental Psychology-Revue Canadienne De Psychologie Experimentale

    (1996)
  • J.M. Gardiner et al.

    Remembering and knowing

  • Cited by (59)

    • Beauty in everyday motion: Electrophysiological correlates of aesthetic preference for human walking

      2022, Neuropsychologia
      Citation Excerpt :

      The regression analyses highlighted the possibility of the association between the higher aesthetic preference for BM and the larger amplitude of the N300 (Fig. 3). Previous ERP studies also reported the association of aesthetic preference for objects (i.e., a product (Shi et al., 2021) and face (Wiese et al., 2014; Zhang and Deng, 2012)) with ERP components, showing a negative deflection at approximately 300 ms after stimulus presentation in the occipitotemporal areas. These findings suggest that aesthetic preference is processed in the later stage of visual perception.

    • Trusting your heart: Long-term memory for bad and good people is influenced by resting vagal tone

      2019, Consciousness and Cognition
      Citation Excerpt :

      In support of this view, the manipulation of specific personally-relevant goals also changes the ability to detect facial cues associated with trustworthiness, an ability that increases when, for example, self-protection motives are activated (Young, Slepian, & Sacco, 2015). This memory advantage for relevant faces is consistent with research on the role of explicit emotional expression and of facial properties that lead to relevant social inferences (trustworthy- and untrustworthy-looking faces (Mattarozzi, Todorov, &Codispoti, 2015; D’Argembeau& Van Der Linden, 2007; Jackson, Wu, Linden, & Raymond, 2009; Wiese, Altmann, & Schweinberger, 2014). Interestingly, some studies have found a memory advantage for negatively valenced faces over positive ones that has not been observed in the current study (Johansson, Mecklinger, &Treese, 2004; Mattarozzi et al., 2015; Lischke, Junge, Hamm, &Weymar, 2018).

    View all citing articles on Scopus
    1

    These authors contributed equally to this work.

    View full text