Time course of attentional bias for fear-relevant pictures in spider-fearful individuals
Introduction
There has been extensive research into attentional biases in anxiety because cognitive theories propose that such biases play a crucial role in the aetiology and/or maintenance of anxiety states (e.g. Eysenck, 1992; Mogg & Bradley, 1998; Williams, Watts, MacLeod, & Mathews (1988), Williams, Watts, MacLeod, & Mathews (1997)). Indeed, there is evidence that generally anxious individuals preferentially attend to aversive stimuli, such as threat-related words and angry faces (e.g. reviews by Mogg & Bradley, 1998; Williams et al., 1997). Research into attentional biases has also been extended to the study of phobias. Öhman (1996) proposed that, as a result of the operation of evolutionarily shaped cognitive mechanisms, individuals have their attention automatically captured by fear-relevant stimuli, and that such mechanisms play a critical role in facilitating defensive action to deal with the feared object or situation. He further proposed that attentional biases should be primarily evident for biologically salient stimuli that are commonly responsible for eliciting phobic states, such as snakes or spiders.
Several studies of attentional biases in specific phobia have used the modified Stroop task, in which participants are required to name the colours of words as quickly as possible while ignoring their content. Results typically show that individuals who are fearful of spiders or snakes show greater interference in colour-naming stimuli which are related to their fears, relative to neutral control stimuli; which is commonly interpreted as vigilance for fear-relevant information. This interference effect has been found for both word and pictorial fear-related stimuli (e.g. Constantine, McNally, & Hornig, 2001; Kindt & Brosschot, 1997; Lavy & van den Hout, 1993; Watts, McKenna, Sharrock, & Trezise, 1986) and this bias is reduced by treatment (Lavy, van den Hout, & Arntz, 1993; van den Hout, Tenney, Huygens, & de Jong, 1997). However, concerns have been raised about the ecological validity of the modified Stroop task (Thorpe & Salkovkis, 1997) and its specificity to threat versus positive information (Constantine et al., 2001; Martin, Williams, & Clark, 1991). Moreover, the interpretation of emotional Stroop interference effects has been questioned, as such effects may not necessarily reflect an attentional bias for threat, but may instead reflect an attempt to avoid processing the aversive information (de Ruiter & Brosschot, 1994).
Other tasks, which have been used to assess cognitive biases in spider phobia, have produced mixed results. Several studies, using a variety of attentional tasks with word and pictorial stimuli, have failed to find evidence of enhanced vigilance for spider-relevant stimuli in spider-fearful individuals, relative to non-fearful controls (Hermans, Vansteenwegen, & Eelen, 1999; Merckelbach, Kenemans, Dijkstra, & Schouten, 1993; Tolin, Lohr, Lee, & Sawchuk, 1999; Wenzel & Holt, 1999). Moreover, two studies have found evidence of avoidance. For example, Tolin et al. (1999) reported that spider-fearful individuals spent relatively less time viewing spider pictures than control pictures. In an eye movement study, Hermans et al. (1999) found that, after initial orienting, spider-fearful individuals were more likely to avert their gaze from spiders, compared with controls. Öhman, Flykt and Esteves (2001) found, using a visual search task, enhanced vigilance for pictures of feared stimuli in a mixed sample of spider-fearful and snake-fearful individuals.
Thus, some studies suggest enhanced attention to fear-relevant stimuli in spider-fearful individuals, while others suggest avoidance. These findings are not necessarily incompatible, because there may be an initial automatic capture of attention by fear-relevant stimuli in phobic individuals, which is rapidly followed by avoidant strategies. Such a ‘vigilant–avoidant’ pattern of processing would seem compatible with recent cognitive models of fear and anxiety (e.g. Mogg & Bradley, 1998; Öhman, 1996). For example, Öhman's (1996) model of fear proposes that the automatic capture of attention by feared stimuli is a key feature of an innate defence system, which allows rapid identification of potential threats. However, the cardinal function of this system is to prompt avoidance of, or escape from, stimuli that may endanger the organism. If so, individuals would have their attention initially captured by a fear-relevant stimulus, but would not necessarily maintain their attention on it. Instead, they may subsequently show rapid disengagement of attention from, and avoidance of, feared stimuli, following initial attentional capture.
A vigilant–avoidant pattern of attentional bias may also be of clinical significance, as it could result in greater detection of threat in the environment in the absence of prolonged exposure (e.g. Mogg & Bradley, 1998). Thus, after initial orienting to threat, the maintenance of attention to threat cues may be subject to conflicting response tendencies, i.e. vigilant monitoring of threat versus avoidance responses that serve to reduce anxious mood and/or danger (e.g. gaze avoidance, escape behaviours). A vigilant–avoidant pattern of bias might be important in maintaining anxiety disorders, because it may result in repeated brief exposures to fear-provoking stimuli without allowing habituation (Marks, 1987). The fear-provoking properties of threat cues may also remain high due to a failure in emotional processing as a result of excessive avoidance (Rachman, 1980).
A limitation of the modified Stroop task is that it cannot examine whether the direction of the attentional bias varies over time. However, the visual probe task has been used to examine the time course of attentional biases (e.g. Bradley, Mogg, Falla, & Hamilton, 1998; Mogg, Bradley, de Bono, & Painter, 1997). In this task, participants respond as quickly as possible to probes that appear following pairs of stimuli (e.g. a threat and a non-threat word). An attentional bias for threat is indicated by faster response times to probes that replace the threat item of a stimulus pair, rather than the non-threat item (e.g. MacLeod, Mathews, & Tata, 1986). The duration of the stimuli can be varied to examine whether the attentional bias occurs very rapidly, consistent with a bias in initial orienting (e.g. stimulus durations of 200 ms or less), and whether it is apparent at longer stimulus durations, which would suggest a bias in maintained attention (e.g. durations of 2000 ms or more). Two studies using this task have suggested initial vigilance, but no subsequent avoidance of threat cues in anxious individuals (Bradley et al., 1998; Mogg et al., 1997). However, such avoidance might be revealed for stimuli that are more fear-provoking than the threat words or angry faces used in these studies.
Hence, the aim of the present study was to investigate the time course of the attentional bias for fear-relevant stimuli in individuals with high levels of spider fear. On each trial, a pair of pictures is presented simultaneously and, on critical trials, one picture depicts a spider and the other a cat (which serves as the control, non-threat stimulus on each trial). The picture pairs are shown for 200, 500 or 2000 ms, and participants respond as quickly as possible to a probe that replaces one of the pictures. The pictures were also subsequently rated to check that they were anxiety-provoking for the high fear participants. We predicted from Öhman's (1996) model that, due to rapid attentional capture, spider-fear individuals would show enhanced vigilance for spiders at the shortest duration of 200 ms, compared with non-fearful individuals. Thus, our hypotheses were as follows:
- 1.
In the 200 ms condition, the high fear group will be faster to respond to probes replacing spider than control stimuli, compared with the low fear group (i.e. an interaction effect of fear group×spider location×probe location on RTs to probes).
- 2.
The high fear group will show a similar attentional bias for fear-relevant stimuli at the intermediate duration of 500 ms, as this duration has been widely used in to demonstrate anxiety-related vigilance effects (e.g. MacLeod et al., 1986).
- 3.
If fear-relevant stimuli subsequently elicit defensive strategies, spider-fearful individuals will show greater avoidance of spider stimuli at the longer exposure duration of 2000 ms, compared with non-fearful individuals, as this exposure time is long enough to allow multiple shifts of attention and to reveal the influence of strategic processing.
- 4.
If initial orienting to threat is followed by avoidance of feared stimuli, the attentional bias in the high fear group will significantly reduce as a function of the exposure duration of the fear-relevant stimuli.
Section snippets
Participants
Volunteers were recruited largely through advertisements in undergraduate colleges and a student newspaper. Twenty-one volunteers who reported being fearful of spiders and who scored 7 or more on a screening questionnaire (a shortened, 15-item version of the Spider Phobia Questionnaire, SPQ, Watts & Sharrock, 1984), were recruited for the high fear group. An additional 21 individuals were selected for the low fear group, who reported no fear of spiders and who were matched for age and gender
Visual probe task
RT data from trials with errors and filler trials were discarded. RTs greater than 2000 ms and more than 3 sds above each participant's mean, were excluded as outliers. In the high fear group, four participants had unusually variable RTs, and one of them also had almost a quarter of data (23%) missing due to errors. Box-and-whisker plots showed that the RT variance of each of these four participants was an extreme outlier relative to the rest of the sample: variance of each was 5–15 times (or
Discussion
The results from the short stimulus exposure condition (200 ms) are consistent with our primary hypothesis of a rapid, initial attentional bias for fear-related stimuli in spider-fearful individuals. These results are compatible with Öhman et al.'s (2001) findings of enhanced, rapid detection of feared stimuli in phobic individuals on a visual search task, and support Öhman's (1993) evolutionary-derived model of fear processing. Furthermore, the present results indicate that this attentional
Acknowledgements
We thank Isabel Ellory and Daniel Smith for their help with the project, and Laurna Rankin for assisting in the preparation of the manuscript. This research was supported by the Wellcome Trust (Grant Ref. 051076).
References (42)
- et al.
Attentional biases for negative information in induced and naturally occurring dysphoria
Behaviour Research and Therapy
(1997) - et al.
The emotional Stroop interference in anxiety: Attentional bias or cognitive avoidance
Behaviour Research and Therapy
(1994) - et al.
Selective attention evidenced by pictorial and linguistic Stroop tasks
Behaviour Therapy
(1993) - et al.
Attentional bias and spider phobia: Conceptual and clinical issues
Behaviour Research and Therapy
(1993) - et al.
Does anxiety lead to selective processing of threat-related information?
Behaviour Research and Therapy
(1991) - et al.
A cognitive-motivational analysis of anxiety
Behaviour Research and Therapy
(1998) - et al.
Some methodological issues in assessing attentional biases for threatening faces in anxiety: A replication study using a modified version of the probe detection task
Behaviour Research and Therapy
(1999) - et al.
Time course of attentional bias for threat information in non-clinical anxiety
Behaviour Research and Therapy
(1997) - et al.
Hypervigilance-avoidance pattern in spider phobia
Anxiety Disorders
(2005) Emotional processing
Behaviour Research and Therapy
(1980)
Information processing in spider phobics: The Stroop colour naming test may indicate strategic but not automatic attentional bias
Behaviour Research and Therapy
Visual avoidance in specific phobia
Behaviour Research and Therapy
Preconscious processing bias in specific phobia
Behaviour Research and Therapy
Questionnaire dimensions of spider phobia
Behaviour Research and Therapy
An inventory for measuring depression
Archives of General Psychiatry
Attentional bias for threatening facial expressions in anxiety: Manipulation of stimulus duration
Cognition and Emotion
Biases in overt and covert orienting to emotional facial expressions
Cognition and Emotion
Anxiety and attentional bias: State and trait
Cognition and Emotion
Snake fear and the pictorial emotional Stroop paradigm
Cognitive Therapy and Research
Anxiety: The cognitive perspective
Anxiety and cognition: A unified theory
Cited by (128)
Anxiety Shapes Amygdala-Prefrontal Dynamics During Movie Watching
2023, Biological Psychiatry Global Open ScienceInhibitory control moderates the association between fear and attention bias to snakes in middle childhood
2022, Cognitive DevelopmentReducing attention bias in spider fear by manipulating expectancies
2020, Behaviour Research and TherapyD-cycloserine as adjunct to brief computerised CBT for spider fear: Effects on fear, behaviour, and cognitive biases
2020, Journal of Behavior Therapy and Experimental PsychiatryAttentional Bias in specific symmetry and cleaning dimensions of obsessive-compulsive disorder
2020, Journal of Anxiety Disorders