Download Attitudes and Evaluation 1 Attitudes and Evaluation

Attitudes and Evaluation 1
Attitudes and Evaluation: Toward a Component Process Framework
William A. Cunningham1 & Marcia K. Johnson2
1
Department of Psychology, University of Toronto
2
Department of Psychology, Yale University
Draft as of April 12, 2004
Address Correspondence to
William A. Cunningham
Department of Psychology
University of Toronto
100 St George Street
Toronto, ON M5S 3G3
CANADA
phone: 416.978.4243
fax:416.978.4811
ACKNOWLEDGEMENTS
Preparation of this chapter and some of the research described was supported by a grant from the
National Institute of Health (MH 62196). We thank Carol Raye, Philip Zelazo, Kris Preacher,
Norman Farb, and Jay Van Bavel for helpful comments on an early version of this chapter.
Attitudes and Evaluation 2
One of the most exciting developments in social psychology is the recent increase in
interest in integrating theories and methods of social psychology, cognitive psychology, and
cognitive neuroscience. Just as the social cognitive revolution which began in the 1980’s
provided new insights into social processes, an additional level of analysis – identifying the
neural correlates of these processes – promises a more complete understanding of important
constructs such as attitudes, prejudice, and emotional regulation.
In this chapter we review social neuroscience research, examining the evaluative
processes that underlie attitudes. It is difficult to think of a social psychological concept more
central than attitudes (Eagly & Chaiken, 1993). The sense that something is good or bad, positive
or negative, pleasant or unpleasant, or to be approached or avoided is critical to almost any
behavior. Indeed, the processes of evaluation and associated behavioral choice are ubiquitous
though often invisible in daily life. Yet, as is increasingly clear, even simple evaluations are
often the integrated outcome of multiple affective and cognitive component processes.
EVALUATIVE PROCESSES AND ATTITUDES
Attitudes, most simply put, are our likes and dislikes (Bem, 1970). Although such a
definition can conjure images of valence tags (the equivalent of little pluses and minuses)
associated with representations of objects, events, concepts, and so forth, Allport (1935)
highlighted the importance of not defining attitudes as rigid, tightly bound responses to a
particular stimulus. That is, attitudes are not simple if-then rules that operate as an S-R
association between the perception of a stimulus and a specific feeling, opinion, or behavior.
Rather, he suggested that the concept of attitude must include the idea of flexibility and defined
an attitude as "a mental and neural state of readiness, organized through experience, exerting a
directive or dynamic influence upon the individual's response to all objects and situations with
Attitudes and Evaluation 3
which it is related." Allport additionally suggested that the way to achieve dynamic, flexible
attitudes that may include ambivalence, was by "reducing attitudes to small enough
components." Following this line of thinking, by making representations of attitudes small
enough (e.g., breaking a representation-attitude association [R-A] into R1-A1, R2-A2, R3-A3, etc.)
a simple concept of attitude might be retained and combinations of attitude representations
would permit levels of attitude complexity.
We take a similar approach in this chapter, suggesting that attitude processes can be
reduced to more elemental units. That is, we suggest that attitudes are the outcome of multiple
affective and cognitive processes, variously recruited and/or tuned to meet situational and
motivational constraints. Although few attitude theorists would deny the flexible nature of
processes underlying attitudes, there has been a (perhaps only implicit) representation focus in
most of social psychology. Attitudes tend to be treated as representations directly retrieved from
memory in response to perceptual cues. Admittedly, the distinction between representation and
process may be somewhat artificial, but we want to shift the emphasis from attitudes as retrieved
to attitudes as constructed dynamically within particular contexts (e.g., situational, cognitive,
motivational). Thus, attitudes and evaluations, and especially the subjective experience
associated with attitudes, comes from a cognitive/affective system that likely operates in a
dynamic and integrated fashion.
An analogy can be made between the concept of an attitude and the concept of a memory.
According to the source monitoring framework (Johnson, Hashtroudi, & Lindsay, 1993; Johnson
& Raye, 1981) remembering is not simply retrieval of a trace, but rather an attribution about the
nature of a mental experience (see also Jacoby, Kelley & Dywan, 1991). Mental experiences
result from a combination of associative and more constructive processes, and attributions or
Attitudes and Evaluation 4
judgments about them are based on a weighted average of various features or types of
information, including additional information that may be retrieved in the service of monitoring
the origin of the mental experience (e.g., imagination or perception). Furthermore the weights
associated with different features or other types of evidence are affected by current context (e.g.,
motivational state, hypotheses, etc.). The activation of information can occur relatively
automatically or via more reflective processes, and the activated information can be evaluated
relatively automatically or by more reflective processes (i.e., although information that arises
relatively automatically is often evaluated relatively automatically, and information that arises
more reflectively is evaluated more reflectively, it is not necessarily the case). Similarly,
attitudes can be viewed as the result of a weighting of the affective qualities associated with
currently activated information. Some evaluative qualities arise more automatically, and some
with more reflective effort (e.g., Johnson & Multhaup, 1992) and some contribute to the attitude
more automatically and some with more reflection.
THE ATTITUDE SYSTEM
As in the social cognitive tradition that underlies much social neuroscience research, we
assume that in order to understand attitudes, one needs to understand the cognitive architecture
that underlies perception and thought. However, even before fully specifying the component
cognitive processes that give rise to our mental experiences, we can move toward a component
process framework by considering two general ways that components could be defined—in
terms of types of information processed, or in terms of the type of operation engaged. With
respect to type of information, historically in the attitude/emotion domain, stimuli have been
assumed to include two key attributes—valence and intensity. With respect to type of process,
as in cognition in general, in the attitude/emotion domain, the distinction between implicit and
Attitudes and Evaluation 5
explicit processes has been important. Here we consider recent evidence that either or both of
these two simple ways of conceptualizing components of attitudes are reflected in identifiable
neural areas or circuits. In addition, we also briefly consider the relation between attitudes and
emotion, whether there are separate neural correlates of positively and negatively valenced
stimuli, the possibility that the same neural systems flexibly process positive and negative stimuli
depending on motivational goals, and how ambivalence is reflected in brain activity.
Attitudes and Emotion
The psychological study of attitudes has been naturally linked to the study of affect and
emotional processing. Thurstone (1931) provided a parsimonious definition of attitudes as
simply “the affect for or against a psychological object.” Research spanning almost a century has
indicated that our self-perceived bodily states provide a powerful input into our evaluative
judgments (Bem, 1972; Damasio, 1995; James, 1884). For example, extending findings that
electrodermal activity (e.g., the skin’s ability to conduct electricity) varied as a function of the
emotionality of presented stimuli (Smith, 1922), electrodermal activity has been found to vary
with attitude extremity (Dysinger, 1931) and agreement with presented attitude statements
(Dickson & McGinnies, 1966). In addition, other psychophysiological methods have been
developed that are sensitive to attitude valence and extremity, such as facial EMG (e.g., changes
in electrical potentials over particular muscles in the face; Cacioppo, Petty, Losch, & Kim,
1986). More recently, research using event related potentials (ERP) and functional magnetic
resonance imaging (fMRI) has demonstrated links between neural activity in regions associated
with emotion and attitudes (e.g., Cunningham, Johnson, Gatenby et al., 2003; Cunningham,
Raye, & Johnson, 2004).
Arousal and Valence
Attitudes and Evaluation 6
For some time, emotion researchers have found that apparently discrete emotional
experiences can be characterized along two orthogonal dimensions. Although the names of these
dimensions vary from theory to theory, they typically involve a dimension of positivity and
negativity (valence), and a dimension of intensity or arousal (Russell, 1979). This distinction
between valence and arousal has had a long history in social psychology. In their two factor
theory of emotion, Schacter and Singer (1962) suggested that emotions are derived from a
combination of arousal and cognitions used to explain the arousal. In several studies, participants
who were experimentally aroused (e.g., given a dose of epinephrine) reported experiencing
greater emotional responses consistent with their situations, but only when they were unaware of
the effects of the arousal manipulations. Such findings suggest that one first experiences a
visceral autonomic response to a stimulus or situation that is devoid of valence, and then one
cognitively interprets this bodily state and generates a specific positive or negative emotion (see
also Cannon, 1929; James, 1884).
In this conception, arousal is associated with an experienced bodily state and valence
with a cognitive appraisal. Of course, these autonomic and cognitive responses could overlap in
time rather that being serial and they could interact in producing specific emotions and attitudes.
Converging evidence for the distinction between arousal and valence has been provided by
recent functional neuroimaging studies. In three fMRI studies that manipulated or parametrically
analyzed both valence and arousal such that the two could be examined orthogonally, different
brain areas were associated with valence and arousal (Anderson, Christoff, Stappen et al., 2003;
Cunningham et al., 2004; Small, Gregory, Mak et al., 2003). In each of these studies, arousal was
associated with amygdala activation and negative valence was associated with right prefrontal
activation. In addition, Adolphs, Russell, and Tranel (1999) asked both a patient with bilateral
Attitudes and Evaluation 7
amygdala damage and controls to rate the valence and arousal of various stimuli. The amygdala
patient rated stimuli to have the same valences as controls, but rated the emotional intensity of
stimuli differently.
Positive and Negative Substrates?
In addition to evidence dissociating valence and arousal, other evidence suggests that
positive and negative components of valence may also be processed distinctly (Cacioppo &
Bernstein, 1994). For example, Davidson and colleagues have found using EEG that there appear
to be separate systems for approach and avoidance related behavior (Davidson & Irwin, 1999).
Viewing or thinking about negative information results in greater right-sided EEG activity than
viewing or thinking about something positive (Cunningham, Espinet, DeYoung, & Zelazo, in
press; Davidson et al., 1990, Jones & Fox, 1992). Interestingly, anger, which is a negative but
approach-related emotion, is associated with greater left than right EEG activity (Harmon-Jones
& Allen, 1998). Thus, laterality does not appear to be synonymous with positivity or negativity
per se, but rather behavioral tendencies associated with evaluative processing.
In the imaging domain, Sutton et al. (1997) found using PET that viewing negatively
valence pictures was associated with activation in the right orbital frontal cortex and the right
inferior frontal cortex, whereas viewing positively valenced pictures was associated with
activation in the left pre- and post-central gyri. More recently, evidence for right lateralized
processing of negative information has been found using fMRI (Anderson et al., 2003;
Cunningham et al., 2003; Cunningham, Raye, & Johnson, 2004). Specifically, areas of the right
inferior frontal cortex and anterior insula consistently appear to be involved more in processing
negative than positive valenced stimuli. Other studies have found that areas of the orbital frontal
cortex (Anderson et al., 2003; Nitschke, Nelson, Rusch, Fox, Oakes, & Davidson, 2003;
Attitudes and Evaluation 8
Kringelbach, O’Doherty, Rolls, & Andrew, 2003) and basal ganglia (Delgado, Nystrom, Fissell,
Noll, & Fiez, 2000) are involved primarily in the processing of positive affect (see Wager, Phan,
Liberzon, & Taylor, 2003 for a meta-analysis). Although such findings do not necessarily imply
that the processing of positive and negative information is fully dissociated, this suggests that
they may involve at least partially separable circuits.
To the extent that the processing of positive and negative information relies on different
brain regions, an important question to resolve regards the different processing time associated
with each. Predictions based on the idea of a negativity bias propose that negative information
takes priority in processing, both in terms of more rapid responses and greater overall influence
(Cacioppo & Bernstein, 1994; Cacioppo & Gardner, 1999; Ito et al., 1998). Thus, one might
expect negative stimuli to be processed more quickly than positive stimuli in terms of brain
activity. Although several studies have investigated potential differences in response time to
negative and positive stimuli, the answer to this question remains unclear.
Several studies have found that the processing of negative information may occur quite
rapidly in the processing stream. For example, Kawasaki, Kaufman, Damasio et al. (2001) found
that the processing of negative, but not neutral or positive, stimuli occurred 120-160ms after
stimulus presentation in single cell recordings of the human orbital frontal cortex. In addition,
negative stimuli appear to be differentiated from positive stimuli in posterior perception areas as
indexed by the early P1 component (Smith, Cacioppo, Larsen, & Chartrand, 2003). Some have
suggested that negative information is privileged such that it is processed more quickly than
positive information – a temporal negativity bias (Cacioppo & Gartner, 1999). Providing support
for this idea, Carretie, Martin-Loeches, Hinojosa, & Mercado (2001) using
magnetoencephalography (MEG) found that negative stimuli were processed 200ms more
Attitudes and Evaluation 9
quickly than positive stimuli in the OFC. Similarly, some ERP components such as the P200
appear to occur more rapidly to negative than positive stimuli (Carretie, Mercado, Tapia, &
Hinojosa, 2001).Yet, other studies have suggested a primacy for positive stimuli. In a study using
faces with various expressions, Batty and Taylor (2003) found that the N170 component
occurred more quickly to faces with pleasant emotional expressions than negative emotional
expressions.
There are several potential explanations for these discrepancies. First, if positive and
negative information are processed by different brain areas, these separable aspects of
information may be processed at different rates for different processes. Thus, for some processes,
negative information may be processed more quickly, and for others positive information may be
processed more quickly (Cunningham et al., in press). Second, we might expect motivational
goals to affect the time courses of the processing of valence. In other words, it may not be that
positive or negative information is processed more quickly per se, but rather that motivation and
attention may determine whether positive or negative information is processed more quickly
depending on situational and motivational factors. Third, valence may be confounded with some
other variable, such as emotional intensity or motivational significance and differences reflect
this other variable.
Flexible tuning of arousal and/or valence components?
Presumably, increased states of arousal direct attention toward motivationally salient
stimuli in complex environments and prepare an organism for behavior. Because different
stimuli may be deemed “important” at different times, a general arousal or vigilance system that
is itself independent of valence might function most efficiently. This system presumably is
involved in monitoring the environment, detecting potentially relevant changes, and redirecting
Attitudes and Evaluation 10
attention, such that significant emotional stimuli can receive enhanced processing. Either after
attention has been directed, or perhaps in parallel, whether a stimulus is positive or negative is
computed. Consistent with the idea of a functional connection between arousal and attention,
studies have demonstrated similar effects for both positive and negative stimuli in brain areas
associated with attention, such as the anterior cingulate gyrus and visual sensory areas (Schupp,
Junghofer, Weike, & Hamm, 2003).
Recent evidence suggests that chronic differences in tuning for valenced information
(positivity vs. negativity bias) as well as situational variables may direct attention and the
perception of emotional intensity. We (Cunningham, Raye, & Johnson, in press) presented
participants with positively and negatively valenced stimuli during fMRI scanning. After
scanning, participants completed an individual differences measure of their prevention and
promotion focus orientation (i.e., participants indicated whether they were more motivated by
positive or negative stimuli, e.g., see Higgins, 1997). As participants were more promotion
focused, greater activation was observed in the amygdala, anterior cingulate gyrus, and
extrastriate cortex for positive stimuli. As participants were more prevention focused, greater
activation was observed in these same regions for negative stimuli. Thus, amygdala and
attentional brain regions were tuned not toward a particular valence, but rather toward stimuli
that were motivationally important. Similarly, Canli, Sivers, Whitfield, Gotlib, & Gabrieli (2002)
found that amygdala activation for happy faces was correlated with extraversion and Mather,
Canli, English et al. (2004) found that whereas amygdala activation was greater for negative
stimuli for young adults, it was greater for positive stimuli for older adults. Presumably, happy
faces signal an important environmental cue for extraverts (more so than for introverts) and
positive stimuli are more important for older adults than for younger adults, consistent with other
Attitudes and Evaluation 11
evidence of an age shift in the relative salience of negative and positive stimuli (Charles, Mather,
& Carstensen, 2003). Thus, although the motivational significance of stimuli may differ between
people, similar structures are involved in processing these stimuli.
Implicit and Explicit Processes in Attitudes
Most current social cognitive theories of attitude propose that two sets of
processes/systems underlie evaluation (see Chaiken & Trope, 1999). Although these models use
different names for the processes and seek to explain different aspects of attitudes (e.g., structure,
function, persuasion, etc.), they typically propose one system that operates relatively
automatically and effortlessly, and another that requires more cognitive attention or effort
(Chaiken, 1980; Gilbert, Pelham, & Krull, 1988; Fazio, 1990; Greenwald & Banaji, 1995; Petty
& Cacioppo, 1984). Often the second, more effortful, system is proposed to play a corrective role
in attitudinal processing by updating or modifying an initial response or judgment deemed to be
inappropriate or suboptimal given current motivational and situational constraints (Devine, 1989;
Fazio, 1990; Petty & Wegener, 1993). Following this logic, dominant models of attitude have
suggested that attitudes reflect dissociated memory representations (Greenwald & Banaji, 1995).
That is, automatic processes activate implicit attitudes and controlled processes activate explicit
attitudes.
We suggest that thinking about attitudes as having both implicit and explicit components
is useful as a heuristic, but attitudes very likely do not break down into a simple implicit/explicit
dichotomy. At a coarse-grained level of analysis, some processes can be thought of as being
relatively automatic, reflexive, and not initiated consciously. Others involve more conscious
deliberation, and usually require more time to initiate. But some “implicit” processes are more
automatic than others, and some “explicit” processes require more deliberation than others. Thus,
Attitudes and Evaluation 12
we suggest that implicit/explicit or automatic/controlled are not dichotomous categories and that
they likely do not operate in an all-or-none fashion. Different explicit (or implicit) evaluations
may recruit different processes and brain systems. Moreover, we suggest that these implicit and
explicit processes begin to interact or become integrated throughout the processing stream. For
example, as time increases between the initiation of evaluative processing and the response we
measure, additional component processes can be engaged that may provide a richer and more
elaborate attitude. In fact, some researchers suggest that at least some attentional processing is
necessary for emotional processing to initiate (Pessoa & Ungerleider, 2005). Thus, although we
refer to the relative automaticity of processes for the purposes of description, we endorse neither
a rigid dichotomy nor a single-factor continuum.
Early/Unaware emotional processing . Given the importance of arousal in directing attention,
it is not surprising that these computations occur at a very early stage of information processing.
In a study of face perception, Asley, Vuilleumier, & Swick (2004) found that ERP signals
differed between emotional and non-emotional faces as early as 120 to 160ms after stimulus
presentation (see also Pizzagalli et al., 2002; Eimer & Holmes, 2002). In fact, ERP differences to
emotional stimuli compared with neutral stimuli have been observed as early as 94ms after
stimulus presentation in occipital regions (Batty & Taylor, 2003). What is particularly interesting
about this rapid emotional processing is that structural processing and identification of facial
features is thought to not occur until 170ms after stimulus presentation (Sagiv & Bentin, 2001),
suggesting that the processing of emotional expression, a signal which can denote safety or
danger, may occur in parallel with facial structural encoding processes. In other words,
emotional significance may be processed before a stimulus has been fully identified (e.g., see
Niedenthal & Kitayama, 1994).
Attitudes and Evaluation 13
Moreover, not only are these early emotional processes relatively automatic, they also
may occur in the absence of conscious awareness. Several studies have now demonstrated
psychological signs of emotional processing to stimuli that participants do not report even having
seen. For example, skin conductance and ERP signals have been detected to the subliminal
presentation of emotional faces as rapidly as 100ms after stimulus presentation (Ohman &
Soares, 1996; Williams, Liddel, Rathjen, et al., 2003). fMRI studies have found amygdala
activation to subliminal presentations of fearful faces (Whalen et al., 1998) and angry faces that
have been previously associated with an aversive stimulus (Morris et al., 1996).
More complex emotional processing. Because some aspects of emotional intensity can occur
relatively automatically does not mean that all of emotional processing is automatic,
unconscious, or inevitable. As more time is available for evaluation, additional processing
components can be brought to bear. Explicit evaluation may induce both more cognitively
complex attitudes and more affectively complex attitudes. For example, simple emotions (e.g.,
fear, joy) can arise from relatively automatic processing of a stimulus, whereas more complex
emotions (e.g., remorse, jealously, empathy) typically involve more reflective processing (e.g.,
Johnson & Multhaup, 1992). Cunningham, Raye, & Johnson (2004) found that some brain
regions (e.g., an area of the orbital frontal cortex and the temporal pole) were associated with
emotional intensity only when participants were making a reflective evaluation. Thus, the
emotional qualities accompanying an evaluative judgment made with or without reflection may
be qualitatively distinct. For example, for some people, affirmative action may elicit relatively
simple feelings of fear or anger arising from automatic processing, and guilt, jealousy, hope or
other complex emotions arising with reflective processing.
Attitudes and Evaluation 14
Additional evidence for differences in the functional roles of emotional processes (e.g.,
gut responses versus more cognitive aspects) in evaluation comes from work by Bechara and
colleagues using the Iowa Gambing Task (see Bechara, 2004 for a review). In this task,
participants select cards from different decks of cards with different reward contingencies. Some
decks provide high immediate gains, but overall are disadvantageous (providing, on average,
fewer rewards), whereas others provide small rewards, but overall are advantageous (on average
providing higher rewards). Although orbital frontal patients showed normal skin conductance
responses (SCRs) when receiving rewards and punishments, they did not show anticipatory SCR
prior to their decisions. This dissociation suggests that some emotional processes are involved in
the planning and simulation of emotional experience to make a decision, and others are involved
in the processing of current rewards and punishments.
Ambivalence and Control. Explicit evaluation involves monitoring, manipulating,
inhibiting, controlling, or differentially weighting evaluative information to reach an evaluative
judgment. When contrasting tasks that require evaluative and nonevaluative judgments, we find
greater activation in medial areas of the prefrontal cortex (PFC) when participants make
evaluative judgments, and lateral areas of the PFC when participants make nonevaluative
judgments (Cunningham et al., 2003, 2004). More importantly, several of the prefrontal regions
were associated with the processing of attitude complexity (having simultaneous positive and
negative responses) or control of an initial response. The functional role of more
reflective/controlled evaluative processes may be to respond to and mitigate evaluative
complexity by withholding responses until more information is available, integrating multiple
sources of information, and/or deliberately retrieving additional memory representations.
Attitudes and Evaluation 15
Such processes are particularly important when stimuli are more complex; for example,
when both positive and negative aspects are activated simultaneously (perhaps by different
subsystems). This simultaneous state of feeling both positive and negative, or ambivalence, has
been shown behaviorally when people win or lose at a task but do not win or lose as much as
they could have; for example, when they only win $5 when $9 was possible (Larsen, McGraw,
Mellers, & Cacioppo, 2004). At the neural level, an area of right orbital frontal cortex (BA47)
shows greater activation when participants make Good/Bad judgments about ambivalent
compared to non-ambivalent (e.g., neutral, clearly positive, or clearly negative) stimuli
(Cunningham et al., 2003; Cunningham et al., 2004).
In such cases of attitude conflict, it is necessary to engage in control processes to reduce
the cognitive or affective inconsistency that ambivalence brings (Festinger, 1954; Heider, 1946).
As in monitoring memories (e.g., Johnson, Hashtroudi, & Lindsay, 1993), the functional role of
more reflective or controlled processes in evaluation may be to withhold responding until more
information is available, integrate multiple sources or dimensions of information, and/or retrieve
additional information. Through these additional control processes, evaluations can be brought
into line with personal values, motivational goals, or situational constraints.
Brain regions that are involved in the regulation of affective states, including areas of the
anterior PFC and anterior cingulate gyrus (Cunningham et al. 2004; Ochsner et al., 2002) are
similar to those involved in cognitive control more generally (Cohen, Botvinick, Carter, 2000;
MacDonald, Cohen, Stenger, & Carter, 2000; Johnson, Raye, Mitchell et al., in press).
Interestingly, several of these regions are involved in control whether one is downregulating
(dampening) or upregulating (enhancing) emotional experience (Ochsner, Ray, Cooper et al., in
press). Furthermore, the degree of prefrontal EEG asymmetry is correlated with a decreased
Attitudes and Evaluation 16
startle response to evocative stimuli, suggesting a direct role of the lateral PFC in the control of
emotional states (Jackson, Mueller, Dolski, Dalton et al., 2003).
As in processes associated with arousal or valence, the processes associated with
ambivalence and control are likely to occur with different temporal dynamics or at different
points in the processing stream, with at least two sets of processes involved in the processing
attitudinal ambivalence. With respect to cognitive conflict and control, Botvinick, Braver, Barch,
Carter, and Cohen (2001) propose that there are separable conflict-detection and regulatory
systems. Whereas the conflict-detection system is proposed to operate relatively automatically,
the regulation system requires or is associated with more controlled processing. Providing
support for this proposed distinction in the domain of evaluation, Cunningham, Raye, & Johnson
(2004) found that some brain areas, such as ventrolateral PFC, correlated with self-reported
ratings of control to particular stimuli, even when making nonevaluative judgments. Although
this may suggest that these processes are automatic, when participants made explicitly evaluative
Good/Bad judgments, these regions were significantly more active. In addition, other brain
regions were active only when participants had an agenda to evaluate, such as areas in anterior
PFC. This suggests that ambivalence, and the control induced by ambivalence, may be
hierarchically organized, such that different degrees of reflection are necessary for different
processes of control. Consistent with this idea, Amodio, Harmon-Jones, Devine, Curtin, Hartley,
& Covert (2004) found ERP signals associated with control that occurred less than 200ms after
the presentation of Black and White faces. Yet, despite signals that control was engaged,
participants on average still showed evidence of prejudice in that they were more likely to
misperceive a tool as a gun when it was preceded by a Black than a White face, suggesting that
more time or additional processes are necessary for more complete control.
Attitudes and Evaluation 17
APPLICATIONS FOR PREJUDICE RESEARCH
An early focus for the social neuroscience study of attitudes has been the important
domain of prejudice. Initial studies demonstrated a role for the amygdala in the processing of
other-race faces. Hart et al. (2000) demonstrated that amygdala activation to supraliminal Black
faces habituated more slowly than White faces for White participants, and the reverse pattern
was found for Black participants. They concluded that all faces are processed immediately for
their threat value, but that in-group faces are deemed safe more quickly than out-group faces.
The role of the amygdala in inter-group perception was further demonstrated by Phelps,
O’Connor, Cunningham et al. (2000) who showed that greater amygdala activation to Black than
White faces was correlated with an indirect measure of race bias that reflects the extent to which
Black is associated with Bad and White with Good, the race IAT. Interestingly, neither of these
studies showed overall greater amygdala activation to Black faces relative to White faces for
White participants.
One potential explanation for not finding the expected greater amygdala activation to
Black than White faces is that control processes may inhibit or reconstrue an activated emotional
response. That is, rather than automatic attitude being separate from controlled aspects of
attitude, the two systems may be more dynamically linked. Thus, for participants viewing long
blocks of Black or White faces (as in Hart et al., 2000 and Phelps et al., 2000), control processes
may dampen any automatic effects that would otherwise be observed. Consistent with this
hypothesis, Cunningham, Johnson, Raye et al. (2004) found that 12 out of 13 White participants
had greater activation to Black than to White faces (which were randomly intermixed), but only
when the faces were presented briefly and masked such that participants did not report seeing the
faces. As one might expect given our control hypothesis, for faces that could be clearly seen, the
Attitudes and Evaluation 18
decreased activation to Black relative to White faces was accompanied by activation in areas of
the PFC and the anterior cingulate gyrus – areas that are associated with cognitive control.
Interestingly, it appears that mental activities that counteract prejudiced thoughts may
diminish control in other situations. According to Baumester, Bratslavsky, Muraven, and Tice
(1998), regulation is a limited resource, and any act of control not only uses up resources at the
time of control, but also for some time afterward while the system recuperates. Richeson and
Shelton (2003) found that after nonprejudiced White participants interacted with a Black
individual, they subsequently performed worse at the Stroop task, a task that requires cognitive
control for incompatible trials (e.g., reporting that the word green is in a red print color). In a
followup fMRI study, Richeson, Baird, Gordon et al. (2003) scanned White participants while
they viewed Black and White faces using fMRI. Afterward, participants performed the Stroop
task. As in Cunningham et al. (2004), greater activation was observed in the right lateral PFC
while participants viewed Black compared with White faces. Furthermore, the degree of right
PFC activity while viewing Black faces during the fMRI task predicted subsequent Stroop
performance, with those with the most right PFC activity performing the worst later. Presumably,
the cognitive cost of control was manifested in the subsequent cognitive task. This pattern of
findings provides support for the idea that nonprejudiced participants attempt to regulate their
emotional responses to Black faces.
It should be noted that the amygdala should not be considered the source of prejudice, but
rather one component that may contribute to biased responses in the context of a larger attitude
system. For example, Phelps, Cannistraci, and Cunningham (2003) reported on a patient who,
despite bilateral amygdala damage, still showed evidence of automatic race biases on an indirect
measure of automatic associations, suggesting that automatic evaluative responses are possible
Attitudes and Evaluation 19
without an amygdala. This should not necessarily be surprising as amygdala activation seems to
now be better characterized as processing emotionally significant stimuli rather than simply
negative, fearful, or threatening stimuli (Canli et al., 2004; Cunningham et al., 2004; Mather et
al., 2004). In the processing of social groups, and people or objects in general, other areas are
associated with processing emotional intensity and valence, notably, right prefrontal and orbital
frontal cortex. Additional patient work examining which aspects of evaluative processing are
impaired with particular forms of damage would be informative.
CONCLUSIONS
Evaluations arise from multiple component cognitive and affective processes that work in
concert to make judgments about the world. Any given process does not typically work alone in
an all or none fashion, but rather various combinations of processes generate qualitatively
different evaluative outcomes. Different situational and motivational constraints, including
whether or not an explicit evaluation is required, activate different component processes,
resulting in different overall evaluations of a stimulus. Thus, the same stimulus can produce quite
different subjective evaluative experiences (for example, a positive attitude in one circumstance
and a negative in another). Attitudes can be thought of as the constructed output of combinations
of currently active component processes.
Already, neuroimaging results suggest that different patterns of brain activity are
associated with emotional intensity, positivity, negativity, ambivalence, and control, and that
activity in different areas may have different temporal signatures. Furthermore, compared to
implicit expression of attitudes, having an agenda to explicitly evaluate a stimulus results in
additional brain areas associated with the processing of each of these aspects. Our evolving
component process framework assumes that implicit and explicit evaluation are not wholly
Attitudes and Evaluation 20
independent; common processes may be involved both in the formation of quick-and-crude
automatic judgments as well as more systematically derived reflective judgments. Future
research is needed to understand how these basic processes work in more dynamic and integrated
ways to construct attitudinal states.
A challenge for future research is to understand how disparate elements or processes
become unified to give rise to phenomenal experience and behavior. It is likely that prefrontal
regions, such as areas in medial PFC that receive inputs from throughout the brain, function as
polymodal integration centers. To the extent that different processes unfold at different rates, this
means that evaluations may vary from second to second (or millisecond to millisecond) as new
information is processed. For example, if a judgment is required early in the processing stream,
an evaluation can be based only on relatively automatic processing of simple emotional
responses (initial arousal and/or valence). More reflective processing of these aspects typically
requires time to initiate, as does regulation and control. Interestingly, with practice, regulation
and control can be accomplished via more automatic processes (e.g., Gollwitzer, Wasel, &
Schaal, 1999). Attitudes, in this framework, reflect the current processing of an integrated
information processing system at any given time. At some point, a judgment may be required,
and this state is the person’s attitude at that given moment. The particular ways in which
information is constrained, weighed, and integrated as attitudes are constructed on line is a
challenging problem. Also of particular interest are the conditions under which the result is not a
unified evaluation—the conditions under which ambivalence persists.
Although understanding attitudes presents a considerable challenge for analysis, there is a
rich history of relevant theoretical ideas and findings from social and cognitive psychology and
intriguing new findings from social cognitive neuroscience. Variability in an individual’s
Attitudes and Evaluation 21
attitudes toward the nominally same stimulus may mean that there is more than one “true”
attitude—that different attitudes reflect particular combinations of elements at different times to
satisfy different constraints. As social neuroscience investigations of attitude and evaluation
continue, it is likely that a clearer picture of this dynamic system will emerge yielding better
understanding of how important aspects of attitudes arise from the intersection of emotional and
cognitive processes. Attitudes are evaluative outcomes, reflecting the way that aspects of
experience that are typically labeled as emotional (valence, intensity, ambivalence) arise from
processes that are typically labeled as cognitive (perceptual and reflective). In other words, just
as emotion and cognition (e.g., Johnson & Multhaup, 1992) and motivation and cognition (e.g.,
Johnson & Sherman, 1990), are interactively linked, attitudes are not something apart from the
cognitive, affective, and behavioral processes that contribute to them and that they in turn
contribute to. In this dynamic, integrated system, attitudes and emotions in turn influence
cognitive processes (including in ways that we sometimes label as motivational).
Attitudes and Evaluation 22
REFERENCES
Adolphs, R., Russell, J. A, & Tranel, D. (1999). A role for the human amygdala in recognizing
emotional arousal from unpleasant stimuli. Psychological Science, 10, 167-171.
Allport G. W. (1935). Attitudes. In Handbook of Social Psychology, ed. C Murchison, pp. 798844. Worchester, MA: Clark Univ. Press.
Amodio, D. M., Harmon-Jones, E., Devine, P. G., Curtin, J. J., Hartley, S. L., & Covert, A. E.
(2004). Neural signals for the detection of unintentional race bias. Psychological
Science, 15, 88-93.
Anderson A.K., Christoff K., Stappen I., Panitz D., Ghahremani D.G., Glover G., Gabrieli J.D.,
& Sobel N. (2003). Dissociated neural representations of intensity and valence in human
olfaction. Nature Neuroscience, 6, 196-202.
Asley, V., Vuilleumier, P., & Swick, D. (2004). Time course and specificity of event-related
potentials to emotional expressions. NeuroReport, 15, 211-216.
Baumeister, R.F., Bratslavsky, E., Muraven, M., & Tice, D.M. (1998). Ego depletion: Is the
active self a limited resource? Journal of Personality and Social Psychology, 74, 12521265.
Bechara, A. (2004). The role of emotion in decision-making: Evidence from neurological
patients with orbitofrontal damage. Brain and Cognition, 55, 30-40.
Bechara, A. Damasio, Tranel, & Anderson, 1994
Bem, D.J. (1970). Beliefs, attitudes, and human affairs. Belmont, CA: Brooks/Cole.
Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experimental
social psychology (Vol. 6, pp. 1-62). New York: Academic Press.
Attitudes and Evaluation 23
Batty, M., & Taylor, M.J. (2003). Early processing of the six basic facial emotional expressions.
Cognitive Brain Research, 2003, 17, 613-620.
Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict
monitoring and cognitive control. Psychological Review, 108, 624-652.
Cacioppo, J, T., & Bernston, G. (1994). Relationship between attitudes and evaluative space: A
critical review, with emphasis on the separability of positive and negative substrates.
Psychological Bulletin, 115, 401-423.
Cacioppo, J. T., & Gardner, W. L. (1999). Emotion. Annual Review of Psychology, 50, 191-214
Cacioppo, J. T., Petty, R. E., Losch, M. E., & Kim, H. S. (1986). Electromyographic activity
over facial muscle regions can differentiate the valence and intensity of affective
reactions. Journal of Personality and Social Psychology, 50, 260-268.
Cannon, W. (1929). Bodily Changes in Pain, Hunger, Fear and Rage. New York: Appleton.
Carretie, L., Martin-Loeches, M., Hinojosa, J. A., & Mercado, F. (2001). Emotion and attention
studies through event-related potentials. Journal of Cognitive Neuroscience, 13, 11091128.
Carretie, L., Mercado, F., Tapia, M., & Hinojosa, J. A. (2001). Emotion, attention, and the
‘negativity bias’, studied through event-related potentials. International Journal of
Psychophysiology, 41, 7-85.
Canli, T., Sivers, H., Whitfield, S.L., Gotlib, I.H., & Gabrieli, J.D. (2002). Amygdala response to
happy faces as a function of extraversion. Science, 296, 2191.
Carter, C. S., Braver, T. S., Barch, D. M., Botvinick, M. M., Noll, D., & Cohen, J. D. (1998).
Anterior cingulate cortex, error detection, and the online monitoring of performance.
Science, 280, 747-749.
Attitudes and Evaluation 24
Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source
versus message cues in persuasion. Journal of Personality and Social Psychology, 39,
752-766.
Chaiken, S., & Trope, Y. (1999). Dual-process theories in social psychology. Guilford Press.
Charles, S.T., Mather, M., & Carstensen, L.L. (2003). Aging and emotional memory: The
forgettable nature of negative images for older adults. Journal of Experimental
Psychology: General, 132, 310–324.
Cohen, J.D., Botvinick, M. & Carter, C.S. (2000). Anterior cingulate and prefrontal cortex: who's
in control? Nature Neuroscience, 3, 421-423.
Cunningham, W. A., Espinet, S. D., DeYoung, C. G., & Zelazo, P. D. (in press). Attitudes to the
Right – and Left: Frontal ERP Asymmetries Associated with Stimulus Valence and
Processing Goal. NeuroImage.
Cunningham, W. A, Johnson, M. K., Gatenby, J. C., Gore, J. C., & Banaji, M. R. (2003).
Component Processes of Social Evaluation. Journal of Personality and Social
Psychology, 85, 639-649.
Cunningham, W. A., Johnson, M. K., Raye, C. L., Gatenby, J. C., Gore, J. C., & Banaji, M. R.
(2004). Separable neural components in the processing of Black and White Faces.
Psychological Science, 15, 806-813.
Cunningham, W. A., Raye, C. L., & Johnson, M. K. (2004). Implicit and explicit evaluation:
fMRI correlates of valence, emotional intensity, and control in the processing of attitudes.
Journal of Cognitive Neuroscience, 16, 1717-1729.
Attitudes and Evaluation 25
Cunningham, W. A., Raye, C. L., & Johnson, M. K. (in press). Neural correlates of evaluation
associated with promotion and prevention regulatory focus. Cognitive, Affective, &
Behavioral Neuroscience.
Damasio, A. R. (1994). Descartes’ error: Emotion, reason, and the human brain. New York:
Putnam.
Davidson, R. J., Ekman, P., Saron, C. D., Senulis, J. A., & Friesen, W. V. (1990). Approachwithdrawal and cerebral asymmetry: emotional expression and brain physiology. I.
Journal of Personality and Social Psychology, 58, 330-341.
Davidson, R. J., & Irwin, W. (1999). The functional neuroanatomy of emotion and affective
style. Trends in Cognitive Science, 3, 11–21.
Delgado,M. R., Nystrom,L. E., Fissel,C., Noll,D. C., & Fiez, J. A. (2000). Tracking the
hemodynamic responses to reward and punishment in the striatum. The American
Physiological Society, 3072–3077.
Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components.
Journal of Personality and Social Psychology, 56, 5-18.
Dickson, H. W., & McGinnies, E. (1966). Affectivity in the arousal of attitudes as measured by
galvanic skin response. American Journal of Psychology, 79, 584-587.
Dysinger, D. W. (1931). A comparative study of affective responses by means of the impressive
and expressive methods. Psychological Monographs, 41, 14-31.
Eagly, A.H., & Chaiken, S. (1993). The psychology of attitudes. Forth Worth, TX: Harcourt
Brace Jovanovich.
Eagly, A. H & Chaiken, S. (1998). Attitude structure and function. In Gilbert, D. T., Fiske, S. T.,
Gardner, L. (Eds). The handbook of social psychology, Vol. 1 (4th ed.). 269-232.
Attitudes and Evaluation 26
Eimer, M., & Holmes, A. (2002). An ERP study on the time course of emotional faces
processing. Neuroreport, 13, 427-431.
Festinger, L. (1954). A theory of social comparison processes. Human Relations 7, 117-40
Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as
an integrative framework. In M. P. Zanna (Ed.), Advances in experimental social
psychology (Vol. 23, pp. 75-109). New York: Academic Press.
Gilbert, D. T., Pelham, B. W. & Krull, D. S. (1988). On cognitive busyness: When person
perceivers meet persons perceived. Journal of Personality and Social Psychology, 54,
733-740.
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, self-esteem, and
stereotypes. Psychological Review, 102, 4-27.
Harmon-Jones, E., & Allen, J. J. B. (1998). Anger and frontal brain activity: EEG asymmetry
consistent with approach motivation despite negative affective valence. Journal of
Personality and Social Psychology, 74, 1310–1316.
Hart, A.J., Whalen, P.J., Shin, L.M., McInerney, S.C., Fischer, H., & Rauch, S.L. (2000).
Differential response in the human amygdala to racial outgroup vs. ingroup face stimuli.
NeuroReport, 11, 2351-2355.
Heider, F. (1946). Attitudes and Cognitive Organization. Journal of Psychology 21, 107-112
Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280-1300.
Ito, T. A., Larsen, J. T., Smith, N. K., & Cacioppo, J. T. (1998). Negative information weighs
more heavily on the brain: The negativity bias in evaluative categorizations. Journal of
Personality and Social Psychology, 75, 887-900
Attitudes and Evaluation 27
Jackson, D.C., Mueller, C.J., Dolski, I., Dalton, K.M., Nitschke, J.B., Urry, H.L., Rosenkranz,
M.A., Ryff, C.D., Singer, B.H., & Davidson, R.J. (2003). Now you feel it, now you don't:
Frontal brain electrical asymmetry and individual differences in emotion regulation.
Psychological Science, 14, 612-617.
James, W. (1884). What is an emotion? Mind, 9, 188-205.
Jacoby, L.L., Kelley, C. M, & Dywan, J. (1989). Memory attributions. In H. L. Roediger, &
F.I.M Craik (Eds). Varieties of memory and consciousness: Essays in honour of Endel
Tulving. (pp. 391-422). Hillsdale, NJ: Lawrence Erlbaum Associates
Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological
Bulletin, 114, 3-28.
Johnson, M. K. & Multhaup, K. S. (1992). Emotion and MEM. In S. Christianson (Ed). The
handbook of emotion and memory: Research and theory. (pp. 33-66). Lawrence Erlbaum
Associates: Hillsdale, NJ.
Johnson, M. K., & Raye, C. L. (l98l). Reality monitoring. Psychological Review, 88, 67-85.
Johnson, M. K., Raye, C. L., Mitchell, K. J., Greene, E. J., Cunningham, W. A., & Sanislow, C.
A. (in press). Using fMRI to investigate a component process of reflection: Prefrontal
correlates of refreshing a just activated representation. Cognitive, Affective, & Behavioral
Neuroscience.
Johnson, M. K., & Sherman, S. J. (1990). Constructing and reconstructing the past and the future
in the present. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and
cognition: Foundations of social behavior (Vol. 2, pp. 482-526). New York: Guilford.
Attitudes and Evaluation 28
Jones, N. A., & Fox, N. A. (1992). Electroencephalogram asymmetry during emotionally
evocative films and its relation to positive and negative affectivity. Brain and Cognition,
20, 280-299.
Kawasaki, H., Kaufman, O., Damasio, H., Damasio, A. R., Granner, M., Bakken, H., et al.
(2001). Single-neuron responses to emotional visual stimuli recorded in human ventral
prefrontal cortex. Nature Neuroscience,4, 15–16.
Kringelbach, M. L., O’Doherty, J. O., Rolls, E. T., & Andrew, C. (2003). Activation of the
human orbitofrontal cortex to a liquid food stimulus is correlated with its subjective
pleasantness. Cerebral Cortex, 13, 1064-1071.
Larsen, J. T., McGraw, A. P., Mellers, B. A., & Cacioppo, J. T. (2004). The agony of victory
and the thrill of defeat: Mixed emotional reactions to disappointing wins and relieving
losses. Psychological Science, 15, 325-330.
Liddel, B. J., Williams, L. M., Rathjen, J., Shevrin, H., & Gordon, E. (2004). A temporal
dissociation of subliminal versus supraliminal fear perception: An event-related potential
study. Journal of Cognitive Neuroscience, 16, 479-486.
Mather, M., Canli, T., English, T., Whitfield, S., Wais, P., Ochsner, K., Gabrieli, J. D. E.,
Carstensen, L. L. (2004). Amygdala responses to emotionally valenced stimuli in older
and younger adults. Psychological Science, 15, 259-263.
MacDonald, A.W., Cohen, J.D., Stenger, V.A., & Carter, C.S. (2000). Dissociating the role of
dorsolateral prefrontal cortex and anterior cingulate cortex in cognitive control. Science,
288, 1835-1837.
Morris, J. S., Ohman, A., & Dolan, R. J. (1998). Conscious and Unconscious emotional learning
in the human amygdala. Nature, 393, 467-470.
Attitudes and Evaluation 29
Niedenthal, P.M., & Kitayama, S. (1994). The hearts’ eye: Emotional influences in perception
and attention. San Diego: Academic Press.
Nitschke, J. B., Nelson, E. E., Rusch, B. D., Fox, A. S., Oakes, T. R., & Davidson, R. J. (2003).
Orbitofrontal cortex tracks positive mood in mothers viewing pictures of their newborn
infants. NeuroImage, 21, 583-592.
Ohman, A., & Soares, J. J. F. (1996). Emotional conditioning to masked stimuli: Expectancies
for aversive outcomes following nonrecognized fear-relevant stimuli. Journal of
Experimental Psychology: General, 127, 69-82.
Ochsner, K.N., Bunge, S.A., Gross, J.J., & Gabrieli, J.D.E. (2002). Rethinking feelings: An
fMRI study of the cognitive regulation of emotion. Journal of Cognitive Neuroscience,
14, 1215-1229.
Ochsner, K. N., Ray, R., Cooper, J., Robertson, E., Chopra, S., Gabrieli, J.D.E., & Gross, J. J. (in
press). For better or for worse: Neural systems supporting the cognitive up- and downregulation of negative emotion.
Pessoa, L., & Ungerleider, L.G. (2005). Visual attention and emotional perception. In L. Itti, G.
Rees, and J.K. Tsotsos (Eds.), Neurobiology of attention. San Diego, CA: Elsevier.
Petty, R. E., & Cacioppo, J. T. (1984). The effects of involvement on responses to argument
quantity and quality: Central and peripheral routes to persuasion. Journal of Personality
and Social Psychology, 46, 69-81.
Petty, R. E., & Wegener, D. T. (1993). Flexible correction processes in social judgment:
Correcting for context-induced contrast. Journal of Experimental Social Psychology, 29,
137-165.
Attitudes and Evaluation 30
Phelps, E. A., Cannistraci, C. J., & Cunningham, W. A. (2003). Intact performance on an indirect
measure of face bias following amygdala damage. Neuropsychologia, 41, 203-208.
Phelps E. A., O’Connor K. J., Cunningham, W. A., Funayama, E. S., Gatenby, J. C., Gore, J. C.,
& Banaji, M. R. (2000). Performance on indirect measures of race evaluation predicts
amygdala activation. Journal of Cognitive Neuroscience, 12, 729-738.
Pizzagalli, D. A., Lehmann, D. Hendrick, A. M., Regard, M., Pascual-Marqui, R. D., &
Davidson, R. J. (2002). Affective judgments of faces modulate early activity (~160 ms)
within the fusiform gyri, NeuroImage 16, 663–677.
Richeson, J. A., & Shelton, J. N. (2003). When prejudice does not pay: Effects of interracial
contact on executive function. Psychological Science, 14, 287-290.
Richeson, J.A., Baird, A.A., Gordon, H.L., Heatherton,T.F., Wyland, C.L., Trawalter, S., &
Shelton, J.N. (2003). An fMRI investigation of the impact of interracial contact on
executive function. Nature Neuroscience. 6, 1323-1328.
Russell, J. A. (1979). Affective space is bipolar. Journal of Personality and Social Psychology,
37, 345-356.
Sagiv, N. & Bentin, S. (2001). Structural encoding of human and schematic faces: Holistic and
part-based processes Journal of Cognitive Neuroscience, 13, 937-951.
Schachter, S., & Singer, J. (1962). Cognitive, social and physiological determinants of emotional
state. Psycholological Review, 69, 379-99.
Schupp, H. T., Junghofer, M., Weike, A. I., & Hamm, A. O. (2003). Attention and emotion: an
ERP analysis of facilitated emotional stimulus processing. Neuroreport, 14, 1107-1110.
Attitudes and Evaluation 31
Small, D. M., Gregory, M. D., Mak, Y. E., Gitelman, D., Mesulam, M. M., & Parrish T. (2003).
Dissociation of neural representation of intensity and affective valuation in human
gustation. Neuron. 39, 701-711.
Smith, W. (1922). The measurement of emotion. London: Paul.
Smith, N. K., Cacioppo, J. T., Larsen, J. T., & Chartrand, T. L. (2003). May I have your
attention please: Electrocortical responses to positive and negative stimuli.
Neuropsychologia, 41, 171-183.
Sutton, S. K., Davidson, R. J., Donzella, B., Irwin, W., & Dottl, D. A. (1997). Manipulating
affective state using extended picture presentations. Psychophysiology, 34, 217-226.
Thurstone, L.L. (1931). Measurement of social attitudes. Journal of Abnormal and Social
Psychology, 26, 249-269.
Wager, T. D., Phan, K. L., Liberzon, I., & Taylor, S. F. (2003). Valence, gender, and
lateralization of functional brain anatomy in emotion: A meta-analysis of findings from
neuroimaging. NeuroImage, 19, 513-531.
Whalen, P. J., Rauch, S. L., Etcoff, N. L., McInerney, S. C., Lee, M. B., & Jenike, M. A. (1998).
Masked presentations of emotional facial expressions modulate amygdala activity
without explicit knowledge. Journal of Neuroscience, 18, 411-418.
Williams, L. M., Liddell, B. J., Rathjen, J., Brown, K. J., Gray, J., Phillips, M., Young, A., &
Gordon, E. (2004). Mapping the time course of nonconscious and conscious perception
of fear: An integration of central and peripheral measures. Human Brain Mapping, 21,
64-74.