Elsevier

Cognition

Volume 158, January 2017, Pages 81-89
Cognition

Original Articles
Physical attraction to reliable, low variability nervous systems: Reaction time variability predicts attractiveness

https://doi.org/10.1016/j.cognition.2016.10.012Get rights and content

Abstract

The human face cues a range of important fitness information, which guides mate selection towards desirable others. Given humans’ high investment in the central nervous system (CNS), cues to CNS function should be especially important in social selection. We tested if facial attractiveness preferences are sensitive to the reliability of human nervous system function. Several decades of research suggest an operational measure for CNS reliability is reaction time variability, which is measured by standard deviation of reaction times across trials. Across two experiments, we show that low reaction time variability is associated with facial attractiveness. Moreover, variability in performance made a unique contribution to attractiveness judgements above and beyond both physical health and sex-typicality judgements, which have previously been associated with perceptions of attractiveness. In a third experiment, we empirically estimated the distribution of attractiveness preferences expected by chance and show that the size and direction of our results in Experiments 1 and 2 are statistically unlikely without reference to reaction time variability. We conclude that an operating characteristic of the human nervous system, reliability of information processing, is signalled to others through facial appearance.

Introduction

Theories of mate selection emphasise the role of attractiveness preferences for guiding mate-choice towards high fitness partners. Specifically, traits that are associated with high fitness should be attractive to potential mates because they offer advantages to a partner as well as to future offspring (Gangestad and Scheyd, 2005, Rhodes, 2006). In humans, facial attractiveness preferences have been repeatedly shown to reflect aspects of mate quality. For example, facial symmetry is attractive and indicative of developmental stability and resilience (Simmons, Rhodes, Peters, & Koehler, 2004) and certain levels of facial colouration are attractive and denote healthy blood oxygenation (Stephen, Coetzee, Smith, & Perrett, 2009). The human face thus reflects a range of important fitness information. Given humans’ high investment in the central nervous system (CNS), we would predict that cues to CNS function would be especially important in mate selection. To date, however, there is no evidence that facial appearance specifically reflects the reliability of the CNS; that is, the degree to which the nervous system functions in a consistent manner. Although consistency of behaviour would cue CNS reliability, appearance cues would offer the advantages of rapid assessment for observers as well as rapid signalling for high-fitness senders. We therefore tested if facial attractiveness preferences are sensitive to the reliability of human nervous system function.

A reliable information processor will produce relatively invariant outputs for a specified input (Shannon, 1948). One important constraint on the reliability of an information processor is the amount of endogenous noise. Endogenous noise can be defined as the amount of unpredictable fluctuation across processing operations within a system. Increasing amounts of endogenous noise eventually become an enemy of reliable information processing (Faisal et al., 2008, Shannon, 1948). In other words, given a repeated input, low noise systems will be reliable, in the sense of producing invariant output. In contrast, high noise systems will produce more variable outputs. We hypothesised that if facial appearance reflects CNS reliability, then facial attractiveness should be correlated with the variability of behavioural outputs.

Interest in understanding variability within individuals is not new (Thouless, 1936, Woodrow, 1932), but it has not been widely acknowledged. Over 50 years ago, Fiske and Rice (1955) conducted a systematic review showing that within-person variability – fluctuation in performance across trials or sessions – is not random, but stable, and provides an enduring marker of underlying function. As we describe below, the stable nature of within-person variability and its functional importance have been further supported in the following decades (MacDonald, Nyberg, & Backman, 2006), and is now most frequently assessed by standard deviation in reaction time (RT) across multiple trials (Li, Huxhold, & Schmiedek, 2004). Under this view, rather than reflecting measurement error that should be ignored, consistency of performance is predictive of psychophysiological function (Fiske and Rice, 1955, Thouless, 1936).

Using a variety of RT tasks, evidence has accumulated across cognitive, neurobiological, behavioural and health levels to demonstrate that increased RT variability is associated with reduced functional capacity of the CNS (MacDonald et al., 2006). At a cognitive level, reduced working memory, attention regulation, inhibition, and processing speed have been associated with RT variability (Kofler et al., 2013). At a neurobiological level, higher variability is associated with reduced structural and functional integrity of large-scale brain networks (MacDonald, Li, & Backman, 2009), as well as altered neurotransmitter function (Li & Rieckmann, 2014). In terms of health outcomes, RT variability predicts long-term mental and physical health with healthier individuals showing more consistent performance and those in a diseased state fluctuating more (MacDonald et al., 2008, MacDonald et al., 2006). Also, as human biology deteriorates in older age, performance on a range of tasks becomes more variable (Li et al., 2004). In sum, cognitive and neurobiological systems are more intact, efficient and healthy in individuals with more consistent performance and compromised in individuals with more varied performance. In addition, reaction time variability and its neural underpinnings have been shown to be heritable, using both quantitative (McLoughlin, Palmer, Rijsdijk, & Makeig, 2014) and molecular genetic approaches (Saville et al., 2014, Saville et al., 2015). In sum, a wealth of evidence from a range of methods supports the importance of processing reliability, as assessed by RT variability, as an important correlate to cognitive, neural, and health-related measures. Consequently, a mate-choice preference for low variability would produce indirect benefits through connection to a high-fitness partner.

Although the cognitive, neural and health correlates of RT variability are becoming clearer, the relationship between RT variability and social signalling remains unknown. Given the large investment of the human species in CNS operation, perceptible cues to CNS function would be expected to be identified and exploited. In particular, we predicted that any visual correlates to CNS reliability should be perceived as attractive. To test whether CNS reliability is visible and attractive, we created composite images from a dataset of 230 individuals who had a headshot photo taken and completed an RT task, which involved raising one of two fingers in response to numerical cues (Fig. 1A). In our first experiment, composite images were made of the 15 individuals from the dataset with the most variable (highest standard deviation of reaction time, SDRT) and least variable (lowest SDRT) latency distributions, for men and women separately. These composite images were shown to a new set of observers who were asked to pick which was more attractive and give an attractiveness rating for each face (Fig. 1B). We measured how frequently low SDRT faces were chosen as more attractive than high SDRT faces, as well as the difference in attractiveness ratings between low and high SDRT faces. If nervous system reliability is signalled through the face, then attractiveness judgments should be associated with low RT variability.

Section snippets

Participants

58 participants (29 female, Mage = 20.3 years, SD = 3.2) took part in the experiment. All participants had normal or corrected-to-normal vision and provided written informed consent prior to data collection. The data reported here were obtained under approval from the Research Ethics and Governance Committee of the School of Psychology at Bangor University. One participant completed the discrimination task but not the ratings task (see below for task details). Thus, 57 participants were included in

General discussion

In sum, we show that the more reliable an individual’s nervous system is, as evidenced by consistency of response time performance, the more attractive they appear to others. In evolutionary terms, the human species has invested heavily in the central nervous system. Here we have shown that at least some aspects of this investment are visible: visual facial traits are correlated with the reliability of information processing, and these traits are perceived as attractive. To our knowledge, this

Author contributions

EEB, CWNS, RW and RR designed the experiments. EEB collected the data. EEB, CWNS, RW and RR analysed the data. RR and RW drafted the manuscript. EEB and CWNS made edits to the manuscript.

Acknowledgements

This work was supported by the Economic and Social Research Council (Grant Number: ES/K001884/1 to R.R.).

References (48)

  • D.G. Mitchem et al.

    No relationship between intelligence and facial attractiveness in a large, genetically informative sample

    Evolution and Human Behavior

    (2015)
  • C.W. Saville et al.

    A neural analogue of the worst performance rule: Insights from single-trial event-related potentials

    Intelligence

    (2016)
  • C.W.N. Saville et al.

    Elevated P3b latency variability in carriers of ZNF804A risk allele for psychosis

    NeuroImage

    (2015)
  • C.W.N. Saville et al.

    COMT Val158Met genotype is associated with fluctuations in working memory performance: Converging evidence from behavioural and single-trial P3b measures

    NeuroImage

    (2014)
  • C.W.N. Saville et al.

    On the stability of instability: Optimising the reliability of intra-subject variability of reaction times

    Personality and Individual Differences

    (2011)
  • C.W. Saville et al.

    Is reaction time variability consistent across sensory modalities? Insights from latent variable analysis of single-trial P3b latencies

    Biological Psychology

    (2012)
  • E.E. Butler et al.

    Investigating the relationship between stable personality characteristics and automatic imitation

    PLoS One

    (2015)
  • J. Cohen

    A power primer

    Psychological Bulletin

    (1992)
  • G. Cumming

    Understanding the new statistics: Effect sizes, confidence intervals, and meta-analysis

    (2012)
  • G. Cumming

    The new statistics: Why and how

    Psychological Science

    (2014)
  • M.R. Cunningham

    Measuring the physical in physical attractiveness – Quasi-experiments on the sociobiology of female facial beauty

    Journal of Personality and Social Psychology

    (1986)
  • M.B. Donnellan et al.

    The Mini-IPIP scales: Tiny-yet-effective measures of the big five factors of personality

    Psychological Assessment

    (2006)
  • A.A. Faisal et al.

    Noise in the nervous system

    Nature Reviews Neuroscience

    (2008)
  • D.W. Fiske et al.

    Intra-individual response variability

    Psychological Bulletin

    (1955)
  • Cited by (1)

    View full text