Selective attention for masked and unmasked threatening words in anxiety: Effects of trait anxiety, state anxiety and awareness

We investigated the effects of awareness on selective attention for masked and unmasked verbal threat material using a computerised version of the emotional Stroop. Participants were assigned to the high trait anxious (HTA) and low trait anxious (LTA) groups on the basis of questionnaire scores, and state anxiety was manipulated within participants through the threat of electric shock. To investigate the effects of awareness on responses to threat, the mode of exposure was blocked such that half the participants received masked trials before the unmasked trials, whereas the other half received the reverse order. The results revealed that there was no difference between the HTA and LTA groups in responses to threat for those who received the masked trials before the unmasked trials. However, when unmasked trials were presented before the masked trials HTA individuals were significantly slower to respond to both masked and unmasked threat words compared to the LTA group, and these effects were not further modified by participants' state anxiety status. The results are discussed in terms of the automatic nature of threat processing in anxiety.     

Selective Processing of Threatening Words in Anxiety: The Role of Awareness

In three experiments, high and low trait-anxious individuals were required to classify a centrally located number as odd or even while ignoring spatially separate threat-related or neutral words. It was found that high anxious subjects showed a selective bias to process threatening stimuli when the stimuli were pattern masked after a brief exposure (14msec). However, this automatic processing bias was apparent only under certain conditions: distractors were within focal attention; masked and unmasked distractors were randomly intermixed; or masked trials were presented after unmasked trials. These results suggest that automatic processing biases in high traitanxious individuals are influenced by contextual factors. There was also a suggestion of a qualitative difference in attentional bias between conscious (unmasked) and nonconscious (masked) states, providing evidence for perception without awareness. The theoretical implications of these results are discussed.     

Anxiety sensitivity: the role of conscious awareness and selective attentional bias to physical threat

Selective attentional biases were examined amongst individuals varying in levels of physical anxiety sensitivity. The dot-probe paradigm was used to examine attention towards anxiety symptomatology, social threat and positive words. Stimuli were presented above (unmasked) and below (masked) the level of conscious awareness. High physical anxiety sensitivity was associated with attentional vigilance for anxiety symptomatology words in both unmasked and masked conditions. For positive words, however, those high in anxiety sensitivity were found to avoid such stimuli when they were masked, whereas they exhibited a relative vigilance when unmasked. If the differences between awareness conditions are reliable, then the impact of the automatic vigilance for threat might be modified by conscious attempts to direct attention towards other types of stimuli.     

Selective attention and avoidance on a pictorial cueing task during stress in clinically anxious and depressed participants

Although it is well established that attentional biases exist in anxious populations, the specific components of visual orienting towards and away from emotional stimuli are not well delineated. The present study was designed to examine these processes. We used a modified spatial cueing task to assess the speed of engagement and disengagement from supraliminal and masked pictorial cues depicting threat, dysphoria, or neutral content in 36 clinically anxious, 41 depressed and 41 control participants. Participants were randomly assigned to a stress or neutral condition. During stress, anxious participants were slow to disengage from masked left hemifield pictures depicting threat or dysphoria, but were quick to disengage from supraliminal threat pictures. Information processing in anxious participants during stress was characterized by early selective attention of emotional stimuli, occurring prior to full conscious awareness, followed by effortful avoidance of threat. Depressed participants were distinct from the anxious group, displaying selective attention for stimuli depicting dysphoria, but not threat, during the neutral condition. In sum, attentional biases in clinical populations are associated with difficulties in the disengagement component of visual orienting. Further, a vigilant-avoidant pattern of attentional bias may represent a strategic attempt to compensate for the early activation of a fear response.     

Effects of stress and anxiety on the processing of threat stimuli

Two experiments extended the work of MacLeod and Mathews (1988) and examined whether a cognitive bias for threat information is a function of state or trait anxiety. Color-naming and attention deployment tasks were used to assess the effects of a stress manipulation procedure on attentional responses in high and low trait anxious subjects. Subjects under high stress selectively allocated processing resources toward threat stimuli, irrespective of their trait anxiety level. There was no consistent evidence of a cognitive bias associated with trait anxiety, and the effect of the stress manipulation did not appear to be mediated by state anxiety. It was suggested that trait factors do not modify attentional biases associated with acute stress, but may influence such biases when stress is prolonged.     

Selective attention for masked and unmasked emotionally toned stimuli: Effects of trait anxiety,state anxiety,and test order

We investigated selective attention for masked and unmasked, threat, and positively valenced words, in high trait anxious (HTA) and low trait anxious (LTA) individuals using the emotional Stroop colour‐naming task. State anxiety was varied within participants through the threat of electric shock. To investigate whether the sequencing of the state anxiety manipulation affected colour‐naming latencies, the ordering of the shock threat and shock safe conditions was counterbalanced across participants. The results indicated that the ordering of the state anxiety manipulation moderated masked and unmasked threat bias effects. Specifically, relative to LTA individuals, HTA individuals showed a threat interference effect, but this effect was limited to those who performed under the threat of shock in the later stages of the experiment. Irrespective of exposure mode and state anxiety status, all individuals showed interference for threat in the early stages of the experiment, relative to a threat facilitation effect in the later stages of the experiment. For the unmasked trials alone, the data also revealed a significant threat interference effect for the HTA group relative to the LTA group in the shock threat condition, and this effect was evident irrespective of shock threat order. The results are discussed with respect to the automatic nature of emotional processing in anxiety.     

Selective attention to threat: A test of two cognitive models of anxiety

Two experiments evaluated differential predictions from two cognitive formulations of anxiety. According to one view, attentional biases for threat reflect vulnerability to anxiety; and as threat inputs increase, high trait anxious individuals should become more vigilant, and low trait individuals more avoidant, of threat (Williams, Watts, MacLeod, & Mathews, 1988, 1997). However, according to a “cognitive-motivational” view, trait anxiety influences the appraisal of stimulus threat value, rather than the direction of attentional bias, and both high and low trait anxious individuals should exhibit greater vigilance for high rather than mild threat stimuli (Mogg & Bradley, 1998). To test these predictions, two experiments examined the effect of manipulating stimulus threat value on the direction of attentional bias. The stimuli included high threat and mild threat pictorial scenes presented in a probe detection task. Results from both studies indicated a significant main effect of stimulus threat value on attentional bias, as there was increased vigilance or reduced avoidance of threat, as threat value increased. This effect was found even within low trait anxious individuals, consistent with the “cognitive-motivational” view. Theoretical and clinical implications are discussed.     

Anxiety and the allocation of attention to threat

Using a probe detection technique we have recently demonstrated that anxious subjects consistently deploy attention towards threat-related stimuli, whereas non-anxious controls tend to move attention away from such material (MacLeod, Mathews, & Tata, 1986).

The current study employed the same paradigm but attempted to distinguish the role of trait and state anxiety by testing high- and low-trait students when state anxiety was relatively low (12 weeks before a major examination) and again when it was relatively high (one week before this examination). High-trait subjects alone tended to shift attention towards generally threatening material on both test occasions. Results for examination-related stimuli were more complex. Increased proximity to the examination was associated with an increase in attentional bias towards such threat stimuli in high-trait subjects, but with increased attentional avoidance of such stimuli in low-trait subjects. It is suggested that the attentional response to currently relevant stress-related stimuli may be associated with neither trait nor state anxiety alone, but with an interactive function involving both these variables. These results are discussed in relation to existing models of emotion and cognition, and alternative interpretations of the findings are considered.     

Selective orienting of attention to masked threat faces in social anxiety

The aims of the study were two-fold: to examine whether previous evidence of a pre-attentive bias for masked threat faces in anxious individuals could be replicated, and to assess the relationship between the predicted bias and measures of trait and social anxiety. Pairs of face stimuli were briefly displayed and masked in a modified version of the visual probe task. Results indicated that high anxious individuals were faster to respond to probes occurring in the spatial location of masked threat rather than neutral faces; consistent with their attention being automatically captured by sub-threshold threat cues. Furthermore, this vigilance effect for masked threat faces appeared to be primarily a function of social anxiety and social avoidance, rather than trait anxiety. It was also more apparent when threat faces were presented in the left visual field, suggestive of right hemisphere involvement.     

Attentional biases for angry faces: Relationships to trait anger and anxiety

In two experiments selective attention to angry faces was investigated in relation to trait anger and anxiety. A pictorial emotional Stroop task comparing colour-naming latencies for neutral and angry faces was employed. In Experiment 1 using an unmasked task, individuals scoring high on trait anger showed an attentional bias for angry faces. In Experiment 2, unmasked and masked versions of the task were used. Individuals were selected on low and high trait anxiety, but there was no indication of a relation between attentional bias scores and anxiety. When individuals were subsequently reallocated to groups on the basis of trait anger scores, the high anger group showed an attentional bias for angry faces in the unmasked and the masked task. Results are discussed in relation to recent neurobiological findings from our laboratory, as reflecting an evolutionary-evolved, content-specific response to the facial expression of anger.     

Attention to Positive and Negative Social-Evaluative Words: Investigating the Effects of Social Anxiety,Trait Anxiety and Social Threat

The study investigated how attention to negative (threatening) and positive social-evaluative words is affected by social anxiety, trait anxiety and the expectation of social threat. High and low socially anxious individuals carried out a modified dot-probe task either while expecting to give a speech or under non-threatening conditions. High socially anxious individuals showed no significant attentional bias towards or away from social-evaluative words. This result significantly contrasted with an identical design that showed avoidance of emotional faces in high socially anxious participants drawn from the same population (Mansell et al ., 1999). Participants who expected to give a speech showed less attentional avoidance of negative and positive social-evaluative words. High trait anxiety was associated with selective attention to negative relative to positive social-evaluative words, consistent with earlier findings of attention to threat cues in high trait-anxious individuals. Implications for designing attention tasks and attentional bias across different dimensions of anxiety are discussed.     

Emotional context influences access of visual stimuli to anxious individuals' awareness

Anxiety has been associated with enhanced unconscious processing of threat and attentional biases towards threat. Here, we focused on the phenomenology of perception in anxiety and examined whether threat-related material more readily enters anxious than non-anxious individuals' awareness. In six experiments, we compared the stimulus exposures required for each anxiety group to become objectively or subjectively aware of masked facial stimuli varying in emotional expression. Crucially, target emotion was task irrelevant. We found that high trait-anxiety individuals required less sensory evidence (shorter stimulus exposure times) to become aware of the face targets. This anxiety-based difference was observed for fearful faces in all experiments, but with non-threat faces, it emerged only when these were presented among threatening faces. Our findings suggest a prominent role for affective context in high-anxiety individuals' conscious perception of visual stimuli. Possible mechanisms underlying the influence of context in lowering awareness thresholds in anxious individuals are discussed.     

The time-course of visual threat processing: High trait anxious individuals eventually avert their gaze from angry faces

Several experiments have shown that anxious individuals have an attentional bias towards threat cues. It is also known, however, that exposure to a subjectively threatening but relatively harmless stimulus tends to lead to a reduction in fear. Accordingly, some authors have hypothesised that high trait anxious individuals have a vigilant-avoidant pattern of visual attention to threatening stimuli. In the present study, 52 high trait anxious and 48 low trait anxious subjects were shown pairs of emotional faces, while their direction of gaze was continuously monitored. For 0-1000 ms, both groups were found to view angry faces more than happy faces. For 2000-3000 ms, however, only high trait anxious subjects averted their gaze from angry faces more than they did from happy faces.     

Terror mismanagement: evidence that mortality salience exacerbates attentional bias in social anxiety

Death anxiety is a basic fear underlying a range of psychological conditions, and has been found to increase avoidance in social anxiety. Given that attentional bias is a core feature of social anxiety, the aim of the present study was to examine the impact of mortality salience (MS) on attentional bias in social anxiety. Participants were 36 socially anxious and 37 non-socially anxious individuals, randomly allocated to a MS or control condition. An eye-tracking procedure assessed initial bias towards, and late-stage avoidance of, socially threatening facial expressions. As predicted, socially anxious participants in the MS condition demonstrated significantly more initial bias to social threat than non-socially anxious participants in the MS condition and socially anxious participants in the control condition. However, this effect was not found for late-stage avoidance of social threat. These findings suggest that reminders of death may heighten initial vigilance towards social threat.     

Vigilance for threat: effects of anxiety and defensiveness

Eysenck’s (1997) theory that attentional biases for threat vary as an interactive function of trait anxiety and defensiveness was tested using a visual probe task. Two stimulus exposure conditions were used to explore a secondary issue concerning attentional allocation over time. Results indicated that, among high trait anxious participants, only those with low levels of defensiveness showed vigilance for threatening faces presented for 500 ms. They also showed an attentional preference for neutral faces, relative to happy faces, irrespective of exposure condition. This pattern was reversed in high trait anxious participants with high levels of defensiveness, who showed an attentional bias towards happy faces (relative to neutral faces) under both exposure conditions. The findings are discussed in relation to their implications for (a) the significance of measures of defensiveness for the conceptualization of high trait anxious individuals, and (b) the status of anxiety-related biases at different stages of information processing.     

The role of selective attention in visual awareness of stimulus features: Electrophysiological studies

Attention and awareness are closely related, but the nature of their relationship is unclear. The present study explores the timing and temporal evolution of their interaction with event-related potentials. The participants attended to specific conjunctions of spatial frequency and orientation in masked (unaware) and unmasked (aware) visual stimuli. A correlate of awareness appeared 100-200 msec from stimulus onset similarly to both attended and unattended features. Selection negativity (SN), a correlate of attentional selection, emerged in response to both masked and unmasked stimuli after 200 msec. This double dissociation between correlates of awareness and SN suggests that the electrophysiological processes associated with feature-based attentional selection and visual awareness of features can be dissociated from each other at early stages of processing. In a passive task, requiring no attention to the stimuli, early electrophysiological responses (before 200 msec) related to awareness were attenuated, suggesting that focal attention modulates visual awareness earlier than does selective feature-based attention.     

Selective processing of masked and unmasked verbal threat material in anxiety: Influence of an immediate acute stressor

Attentional biases for threat were investigated using a computerised version of the emotional Stroop task. The study examined the influence of state and trait anxiety by employing a student sample assigned to high trait anxious (HTA; n=32) or low trait anxious (LTA; n=32) groups on the basis of questionnaire scores, and state anxiety was manipulated within participants through the threat of electric shock. Threatening words that were either unrelated (e.g., cancer, danger) or related to the threat of shock (e.g., electrocute, shock) were presented to participants both within and outside of awareness. In the latter condition a backward masking procedure was used to prevent awareness and exposure thresholds between the target and mask were individually set for each participant. For unmasked trials the HTA group showed significant interference in colour naming for all threat words relative to control words when performing under the threat of shock, but not in the shock safe condition. For the masked trials, despite chance performance in being able to identify the lexical status of the items, HTA participants showed facilitated colour naming for all threat words relative to control items when performing under threat of shock, but this effect was not evident in the shock safe condition. Neither valence of the items nor the threat of shock influenced colour naming latencies in either exposure mode for the LTA group.     

Attentional bias in high- and low-anxious individuals: Evidence for threat-induced effects on engagement and disengagement

Attentional bias to threatening visual stimuli (words or pictures) is commonly present in anxious individuals, but not in non-anxious people. There is evidence to show that attentional bias to threat can be induced in all individuals when threat is imposed by threat not of symbolic nature, but by cues that predict aversive stimulation (loud noise or electric shock). However, it is not known whether attentional bias in such situations is still influenced by individual differences in anxiety. This question was addressed in two experiments using a spatial cuing task in which visual cues predicted the occurrence of an aversive event consisting of a loud human scream. Speeded attentional engagement to threat cues was positively correlated with trait anxiety in Experiment 1. Experiment 2 showed that speeded attentional engagement was present only in participants selected for high anxiety but not in low-anxious participants. In both experiments, slower disengagement from threat cues was found in all participants, irrespective of their trait anxiety levels.     

Attentional bias in high- and low-anxious individuals: evidence for threat-induced effects on engagement and disengagement

Attentional bias to threatening visual stimuli (words or pictures) is commonly present in anxious individuals, but not in non-anxious people. There is evidence to show that attentional bias to threat can be induced in all individuals when threat is imposed by threat not of symbolic nature, but by cues that predict aversive stimulation (loud noise or electric shock). However, it is not known whether attentional bias in such situations is still influenced by individual differences in anxiety. This question was addressed in two experiments using a spatial cuing task in which visual cues predicted the occurrence of an aversive event consisting of a loud human scream. Speeded attentional engagement to threat cues was positively correlated with trait anxiety in Experiment 1. Experiment 2 showed that speeded attentional engagement was present only in participants selected for high anxiety but not in low-anxious participants. In both experiments, slower disengagement from threat cues was found in all participants, irrespective of their trait anxiety levels.     

Time-course of attention for threatening pictures in high and low trait anxiety

Cognitive studies about anxiety suggest that the interplay between automatic and strategic biases in attention to threat is related to the persistence of fear. In the present study, the time-course of attention to pictures with varying threat levels was investigated in high trait anxious (HTA, n=21) and low trait anxious (LTA, n=22) students. In a visual probe detection task, high and mild threat pictures were presented at three durations: 100, 500, and 1250 ms. Results indicated that all individuals attended to the high threat pictures for the 100 ms condition. Differential responding between HTA and LTA individuals was found for the 500 ms condition: only HTA individuals showed an attentional bias for mild threatening stimuli. For the 1250 ms condition, the HTA individuals attended away from high and mild threat pictures. The observed pattern of differential attention to threatening pictures may explain the persistence of fear in HTA individuals.