Goal-directed attentional deployment to emotional faces and individual differences in emotional regulation

Abstract

This study investigated whether goal-directed attentional deployment to emotional faces serves as an effective mechanism for emotion regulation and whether individual differences in this ability predicts more effective emotion regulation. Undergraduate participants (N = 109) performed the dot-probe task under stress and were given either a goal to focus their attention on happy faces and avoid angry faces or no attentional goal. Participants given this goal reported nearly three times less frustration in reaction to a stressful anagram task compared to those not given this goal. In addition, those with a greater ability to focus on happy faces and avoid angry faces persisted significantly longer on a stressful anagram task. Trait anxiety did not moderate these effects. These findings have important implications for theories of emotion regulation and anxiety-related attentional biases. Training goal-directed attentional deployment holds considerable promise for future research on developing effective emotion regulation techniques.

Introduction

Attentional deployment toward threatening information is considered fundamental to the etiology and maintenance of nonclinical and clinical anxiety (Mathews & MacLeod, 2005). However, studies of attentional biases tend to focus on automatic attentional deployment rather than goal-directed attention. Bar-Haim, Lamy, Pergamin, Bakermans-Kranenburg, and IJzendoorn (2007) integrated a number of attentional bias theories and proposed that a guided threat-evaluation system can override an automatic threat-evaluation and redirect attention to perform goal-directed behavior. This is compatible with emotion regulation theory that suggests goal-directed attentional deployment is a primary mechanism through which people regulate emotion (Gross & Thompson, 2007). However, these hypotheses remain largely untested and have not been integrated.

Mogg, Bradley, Miles, and Dixon (2004) provided support for the hypothesis that some anxious individuals can override an automatic attentional threat-bias. They gave high-trait-anxious, low-trait-anxious, and blood-fearful individuals a picture version of the dot-probe task. In this task, individuals were first presented two pictures simultaneously, where one was a threatening picture and the other was emotionally-neutral. These pictures were then immediately replaced by a simple probe that appeared behind just one of the pictures. The faster a participant responds to the probe following a threatening picture, compared to a neutral picture, the stronger the attentional bias toward threat. The results showed that both high-trait-anxious and blood-fearful individuals had a bias toward threat early in the time course of attention. In contrast, late in the time course, the threat-bias was no longer present for high anxious individuals and blood-fearful individuals reversed their bias and shifted it away from threat. This study highlights the importance of the time course in attentional processing and suggests that some individuals are able to override an attentional bias toward threat later in the time course of attention (also see Derryberry and Reed, 2002, Garner et al., 2006).

Other evidence related to whether those high in anxiety can override an initial automatic bias toward threat is mixed. For example, MacLeod and Mathews (1988) showed that when state anxiety was high for all participants, high-trait-anxious participants demonstrated an attentional bias toward threat. In contrast, low-trait-anxious participants demonstrated an attentional bias away from threat. This study suggests some individuals are able to override a bias toward threat, and even direct attention away from threat when under stress. In contrast, Mogg, Mathews, Bird, and Macgregor-Morris (1990) had high and low-trait-anxious participants perform a stressful anagram task and showed that all participants, irrespective of anxiety level, exhibited an attentional bias toward threat. This suggests that under stress people exhibit an attentional bias toward threat and do not override this bias. However, neither study investigated the time course of the attentional bias under stress. Therefore, it is unclear whether individuals exhibited an initial bias toward threat, and then later used controlled attention to avoid threat.

Studies of attentional biases focus on anxiety-related attentional biases, rather than on generalized attentional deployment mechanisms (although see Öhman, 2002, Öhman et al., 2001). In contrast, emotion regulation theory develops generalized attentional deployment mechanisms applicable beyond anxiety. Both literatures would benefit from integration. Investigating goal-directed deployment of attention serves as a bridge between the literatures. It is proposed to be both a mechanism by which an anxiety-related attentional bias can be overridden and a generalized mechanism of emotion regulation.

Although limited, there is also evidence that goal-directed attentional deployment serves as a generalized mechanism of emotion regulation. There is emerging evidence in the aging literature that suggests older adults preferentially deploy their attention toward positive stimuli and away from negative stimuli, whereas younger adults do not (Carstensen and Mikels, 2005, Mather and Carstensen, 2005). In a recent study, both younger and older adults demonstrated an initial bias toward emotional pictures. In contrast, only older adults deployed their attention away from threat later in the time course of attention (Rosler et al., 2005). In another recent study, older and younger adults were given pictures of negative-neutral scene pairs (Knight et al., 2007). Older adults deployed their attention away from threat for longer durations than the younger adults. Socioemotional selectivity theory suggests the tendency for older adults to avoid threat and attend to positive stimuli is a goal-directed behavior that contributes to increased emotion regulation ability (Carstensen, 1995, Mather and Carstensen, 2005).

These findings from both the anxiety and aging literatures are consistent with the hypothesis that individuals use goal-directed attentional deployment to regulate emotion. However, none of these studies actually manipulated an emotional goal to preferentially deploy attention toward or away from emotional stimuli. Some recent studies have manipulated an attentional bias away from threat or toward positive stimuli. They have shown this to be an effective way to enhance emotion regulation effectiveness (MacLeod et al., 2002, Wadlinger and Isaacowitz, 2008). However, an explicit emotional goal related to attentional deployment was not given. The previous studies employed the dot-probe paradigm and modified the task so that the location of the probe was completely predictable. Specifically, the probe always followed either a neutral stimulus (as opposed to a threatening stimulus) or a positive stimulus (as opposed to a threatening stimulus) so that participants would learn to avoid threat and focus attention on either neutral stimuli or positive stimuli. Minimal demand was placed on goal-directed attention because no goal was given. To place a significant demand on goal-directed attention, an emotional goal should be given that instructs individuals to focus on a particular emotional category. In addition, the location of the probe should be unpredictable so that individuals must maintain the emotional goal throughout the entire task.

Even if individuals are directly given a goal regarding how attention should be deployed to emotional stimuli, it is possible that they modify their attentional goals depending on current emotion regulation demands. In previous studies very little demand was placed on emotion regulation when attentional deployment mechanisms were being investigated (Carstensen and Mikels, 2005, MacLeod et al., 2002, Mather and Carstensen, 2005, Wadlinger and Isaacowitz, 2008). Therefore, to test whether goal-directed attentional deployment serves as a mechanism for effective emotion regulation, attention deployment mechanisms should be studied when a salient demand is placed on emotion regulation.

In order to determine the effectiveness of goal-directed attentional deployment, the choice of criterion measures of emotion regulation is critical. Subjective measures of emotional reactivity most often serve as the criterion measures of effective emotion regulation. However, objective measures of behavior provide important convergent (or divergent) information regarding the effectiveness of emotion regulation. Baumeister, Vohs, and Funder (2007) observed that while examining inner cognitive and experiential processes are important in psychology, behavior should be just as important, and yet it is rarely measured. In addition, the experiential, physiological, and behavioral outcomes of emotion and emotion regulation are often loosely coupled and sometimes discrepant (Bradley and Lang, 2000, Hubert and de Jong-Meyer, 1990, Mauss et al., 2006, Mauss et al., 2005). Therefore, both self-report and behavioral outcomes should be assessed to gain a more complete picture of emotion regulation effectiveness.

Vogel and Awh (2008) noted that recently cognitive neuroscientists have begun to use individual differences in cognitive processes to constrain cognitive theory. This is an example of Cronbach’s (1957) call for the integration of correlational and experimental psychology – a challenge often heralded but rarely employed. Studies of attentional deployment in emotion regulation theory offer an excellent opportunity to capitalize on individual differences. Therefore, the current study will integrate experimental and individual-difference approaches to determine whether attentional deployment can serve as a mechanism of emotion regulation.

It is predicted that those given the emotional goal to deploy attention toward positive emotional stimuli under stress will exhibit less emotional reactivity to a subsequent stressor (e.g., report lower levels of frustration) and persist longer on a subsequent stressful task. In addition, individual differences in the ability to employ goal-directed attention to positive emotional stimuli under stress will also be predictive in that those with a greater ability to deploy attention will exhibit less emotional reactivity to a subsequent stressor and greater persistence on a subsequent stressful task. The literature is conflicted regarding the role of anxiety in one’s ability to employ goal-directed attention. Anxiety’s role will be investigated by testing whether it moderates individuals’ ability to strategically deploy attention to positive emotional stimuli and away from negative emotional stimuli.