Intact performance on an indirect measure of race bias following amygdala damage

https://doi.org/10.1016/S0028-3932(02)00150-1Get rights and content

Abstract

Recent brain imaging and lesion studies provide converging evidence for amygdala involvement in judgments of fear and trust based on facial expression [Adolphs et al., Nature 393 (1998) 470; Adolphs et al., Neuropsychologia 37 (1999) 1111; Breiter et al., Neuron 17 (1996) 875; Winston et al., Nat. Neurosci. 5 (3) (2002) 277]. Another type of social information apparent in face stimuli is social group membership. Imaging studies have reported amygdala activation to face stimuli of different racial groups [Hart et al., NeuroReport 11 (11) (2000) 2351]. In White American subjects, amygdala activation to Black versus White faces was correlated with indirect, implicit measures of racial evaluation [Phelps et al., J. Cogn. Neurosci. 12 (5) (2000) 729]. To determine if the amygdala plays a critical role in indirect social group evaluation, as suggested by the imaging results, a patient with bilateral amygdala damage and control subjects were given two measures of race bias. All subjects were female, White Americans. The Modern Racism Scale (MRS) is a direct, self-report measure of race attitudes and beliefs. The Implicit Association Test (IAT) is an indirect, automatic evaluation task. Performance on the two tasks did not differ between the patient with amygdala damage and control subjects. All subjects showed a pro-Black bias on the direct, explicit measure of race beliefs, the MRS, and a negative evaluation towards Black faces on the indirect measure of race evaluation, the IAT. These results indicate that even though amygdala activation to Black versus White faces is correlated with performance on indirect measures of race bias [Phelps et al., J. Cogn. Neurosci. 12 (5) (2000) 729], the amygdala is not critical for normal performance on the IAT.

Introduction

The amygdala is a small, almond-shaped structure in the medial temporal lobe that is primarily known for its role in emotional learning and memory [18]. Across a wide range of species, the amygdala has been shown to be necessary for the acquisition and expression of aversive conditioning [10], [17]. It has also been implicated in the modulation of memory with arousal [20]. However, more recently, investigations of the human amygdala have suggested that it may play a role in a limited range of social judgments as well, primarily judgments derived from facial stimuli.

In a series of studies, Adolphs et al. [2] have shown that damage to the amygdala leads to impairments of judgments of fear from facial expressions [29]. These studies have demonstrated that patients with amygdala damage, although they can often identify a fear expression, do not find this expression nearly as “fearful” as do normal controls. Functional imaging studies have provided converging evidence for a specific role of the amygdala in the processing fear facial expressions. A number of studies have demonstrated greater amygdala activation to fear, relative to other facial expressions [7], [21]. This differential amygdala response to fear expressions has also been observed when the faces are presented so quickly subjects are unaware of their presentation, demonstrating that the amygdala’s response is automatic and not dependent on conscious, control processes [27].

More recently both lesion and imaging studies have suggested amygdala involvement in more subtle social judgments. Investigations of patients with amygdala damage have found deficits in the ability to judge whether an individual appears to be approachable or trustworthy [1]. When shown pictures of individuals that normal control subjects rate as appearing untrustworthy or unapproachable, patients with amygdala damage tend to rate them as both trustworthy and approachable. Patients with amygdala lesions fail to pick up on the subtle differences in facial expression or appearance that normal subjects use to judge that an individual may not be trustworthy or friendly. Additional support for amygdala involvement in the complex judgment of trust from faces comes from an fMRI study by Winston et al. [28]. Subjects were asked to view faces and make a judgment of school age (high school or college) or trust (trustworthy or untrustworthy). After scanning, all of the faces were rated for trust. Winston et al. [28] reported greater amygdala activation to those faces rated as untrustworthy. This amygdala activation to untrustworthy faces was found regardless of the task performed while scanning (school or trust) suggesting that the amygdala response is automatic and not dependent on cognitive mediation. There was also a correlation between the trust ratings and facial expression, with faces rated as untrustworthy appearing more sad or angry. These results indicate that the amygdala is involved in the automatic judgment of personal qualities from facial expression.

The studies described above suggest a role for the amygdala in making social judgments about the characteristics of individuals from their facial expression; is the person fearful, trustworthy or approachable. Another variable used to judge personal characteristics from facial stimuli is social group membership. The determination that a person is male or female, young or old, Black or White may result in assumptions concerning that individual’s personal qualities. Recent functional imaging studies examining the neural basis of social group evaluation have indicated amygdala involvement when viewing faces of different racial groups. Two fMRI studies have reported activation of the amygdala in response to the presentation of Black and White faces with neutral facial expressions [15], [24]. In a study by Hart et al. [15], Black and White American subjects viewed pictures of Black and White faces. Overall, there was greater amygdala activation to outgroup faces than ingroup faces. In other words, Black American subjects showed more amygdala activation to White than Black faces, while the White American subjects showed the opposite pattern of results. This study demonstrated amygdala activation to social group membership, but did not indicate any potential behavioral role for this amygdala response.

A study by Phelps et al. [24] with White American subjects showed that the extent of amygdala activation to Black versus White faces was correlated with some indirect behavioral indications of racial group evaluation which were assessed after scanning. There is a growing body of behavioral evidence indicating a dissociation between some indirect (implicit) and direct (explicit) assessments of race evaluation. In spite of a decline over the past several decades of prejudicial attitudes toward Black and White social groups as measured by explicit self-report [6], [25], there is robust evidence of racial bias using indirect assessments that bypass awareness and conscious control [5], [11], [14]. In the Phelps et al. [24] study the White American subjects who showed greater race bias on two indirect measures of race evaluation also showed greater amygdala activation while viewing the Black versus White faces. There was no relation between amygdala activation and race bias as measured by explicit self-report. These results begin to delineate the different neural systems underlying direct and indirect evaluations of racial groups, and suggest that the amygdala may be involved in the automatic, implicit evaluation of social group information derived from faces.

The finding that amygdala activation is related to the indirect evaluation of social group information from facial stimuli is consistent with studies indicating indirect or automatic processing of other types of social information from faces [27], [28]. However, the findings demonstrating a role for the amygdala in fear and trust perception are supported by converging evidence from brain imaging and lesion studies. The demonstration of activation of a particular brain region suggests that this region may be involved in the processing of the stimuli or behavior executed. But imaging data by itself cannot indicate the precise role that a brain region may have in a given task. Lesion studies can indicate if a specific brain region is critical for a given behavior, although like imaging studies there can be difficulties in interpreting lesion results [22]. Combining the two techniques can provide powerful converging evidence that activity in a brain region is not only correlated with stimulus processing, but critical for the execution of behavior.

In the present study we attempt to determine if the amygdala plays a critical role in the indirect evaluation of race bias. The correlation between amygdala activation and indirect measures of race bias suggests that the amygdala is engaging in the automatic processing of social group information from facial stimuli. However, this correlation does not indicate how the amygdala may be involved in the indirect evaluation of racial groups. By examining performance of a patient with bilateral amygdala damage on direct and indirect measures of race bias, we can determine if the amygdala is critical for normal performance on these tasks.

The measures of racial bias used were the same as those reported in the brain imaging study by Phelps et al. [24]. The Implicit Association Test (IAT) [14] was used to indirectly measure race bias. The term “bias” in this context refers to the presence of an indirect or non-controllable behavioral response that exhibits preference for one group over another. The IAT measures the degree to which social groups are automatically associated with positive and negative evaluations. The IAT was followed by the Modern Racism Scale, a commonly used measure of conscious, self-reported beliefs and attitudes toward Black Americans [19].

Section snippets

Subjects

A patient with bilateral damage to the amygdala (SP) was assessed along with two normal control subjects. SP is a 58-year-old White female who, at the age of 48 years, had her right amygdala removed as a result of anteromedial temporal lobe resection for medically intractable epilepsy. Her right temporal lobe resection included partial removal of the middle and inferior temporal gyri, and complete removal of the hippocampus and parahippocampus. Prior to her surgery, an additional lesion was

Modern Racism Scale

SP’s mean score for the items on the MRS was 1.8 and the mean for the two control subject was 1.2 (1.4 and 1). Both of these scores indicate pro-Black beliefs (with 6 being strongly anti-Black and 1 being strongly pro-Black), suggesting that SP and control subjects show a similar bias on this self-report measure of race attitudes.

Implicit Association Test

The measure of bias on the IAT is the discrepancy in reaction time between the Black+good/White+bad trails and the Black+bad/White+good trials. As can be seen in Fig.

Discussion

The patient with bilateral amygdala damage performed similarly to control subjects on both the direct (Modern Racism Scale) and indirect (Implicit Association Test) measures of race evaluation. As in the Phelps et al. [24] imaging study, all the White American subjects in the present study showed a pro-Black bias on the direct, self-report measure of race attitudes and beliefs. They also showed a negative bias towards Black faces on the indirect assessment of race evaluation. Although it

Acknowledgements

The studies were conducted at the Yale University School of Medicine. The authors would like to thank Benjamin Holmes and Laura Thomas for assistance in the preparation of this manuscript. This research was supported by the James S. McDonnell Foundation, 21st Century Scientist Award to EAP.

References (29)

  • R.H. Fazio et al.

    Variability in automatic activation as an unobtrusive measure of racial attitudes: a bona fide pipeline?

    Journal of Personality and Social Psychology

    (1995)
  • I. Gauthier et al.

    Activation of the middle fusiform face area increases with expertise in recognizing novel objects

    Nature Neuroscience

    (1999)
  • A.J. Golby et al.

    Differential responses in the fusiform region to same-race and other-race faces

    Nature Neuroscience

    (2001)
  • A.G. Greenwald et al.

    Measuring individual differences in social cognition: the Implicit Association Test

    Journal of Personality and Social Psychology

    (1998)
  • Cited by (86)

    • The racially diverse affective expression (RADIATE) face stimulus set

      2018, Psychiatry Research
      Citation Excerpt :

      Initial stimulus sets of facial expressions were developed to examine processing and production of emotions (Ekman and Friesen, 1976; Ekman and Friesen, 1978; Russell and Bullock, 1985), however many of the earliest sets pose some potential limitations for researchers due to the narrow number of emotions depicted, number of models, or homogeneity of race and ethnicity (Erwin et al., 1992; Phillips et al., 1998; Winston et al., 2002). To address some of these limitations researchers have continued to create and standardize face stimuli (Hart et al., 2000; Kanade et al., 2000; Phelps et al., 2000; Gur et al., 2002; Batty and Taylor, 2003; Phelps et al., 2003; Tanaka et al., 2004; Pantic et al., 2005; Ashraf et al., 2009; Samuelsson et al., 2012; Dalrymple et al., 2013; Giuliani et al., 2017). The growing use of face stimuli in psychological research has led to the creation of many stimulus sets, each with its own specific attributes (e.g., expanded set of models, ages, photographed angles, emotions, races and ethnicities).

    • Cathodal transcranial direct current stimulation of the extrastriate visual cortex modulates implicit anti-fat bias in male, but not female, participants

      2017, Neuroscience
      Citation Excerpt :

      Indeed, it has been previously reported that the frontal cortex, anterior temporal lobes and the limbic system are key areas implicated in the forming of social prejudice. More specifically, the amygdala has been found to be critically involved in cognitive and affective learning, including implicit attitudes (Amodio and Devine, 2006; Dolan et al., 2000; Phelps et al., 2003; Stanley et al., 2008). Furthermore, recent experimental evidence has proposed a critical involvement of the anterior temporal lobes in expressing prejudice by means of conceptual processing (Snyder et al., 2004; Gallate et al., 2011).

    • The Neural Mechanisms of Prejudice Intervention

      2016, Neuroimaging Personality, Social Cognition, and Character
    • Attitudes

      2015, Brain Mapping: An Encyclopedic Reference
    View all citing articles on Scopus
    View full text