Abstract
We read the paper by Zhang et al. 1 with great interest as it addresses a timely issue in advancing current PTSD treatments. While existing therapies benefit many patients, up to 30% do not fully recover and continue to experience residual symptoms impacting daily functioning.2,3 Therapies targeting attentional biases (AB) have emerged as promising adjuncts, drawing from cognitive models that highlight the role of these biases in the onset and persistence of the disorder. 4
Zhang et al. conducted a meta-analysis to evaluate the effectiveness of attentional bias modification (ABM) versus attention control training (ACT). ABM aims to reduce negative attentional biases by training individuals to focus on nonthreatening stimuli, while ACT allows individuals to maintain their natural attentional patterns with threat-related contingencies being irrelevant to task performance. Zhang et al. concluded that ACT was more effective in treating PTSD symptoms, underscoring its therapeutic potential.
Beyond the limitations acknowledged by Zhang et al., such as the limited number of included studies and significant heterogeneity between them, we identify three major issues that question the therapeutic potential of ACT:
First, the studies included in the meta-analysis did not use a genuine control condition, preventing definitive conclusions about the effectiveness of either ABM or ACT. Studies incorporating alternate non-emotional tasks report no difference in PTSD symptom reduction compared to ABM and ACT. 5 This suggests that non-specific attentional control effects may be at play and should be properly controlled by using an inactive control condition, such as a waiting list, to account for changes unrelated to these techniques, including treatment-related expectancy effects.
Second, Zhang et al. did not observe a significant group difference between ABM and ACT in reducing AB, raising questions about the mechanism through which ACT might exert its therapeutic effects. Moreover, few studies assessed whether symptom reduction correlates with a decrease in AB, making it impossible to ascertain the association between changes in AB and symptom reduction.
Third, a broader concern exists regarding the very presence of AB in PTSD. A meta-analysis that included dot-probe data from clinically anxious individuals and individuals with PTSD across 13 ABM RCTs found no significant evidence that these individuals are characterized by a threat-related attention bias at baseline. 6 These findings question the assumption that AB toward threats is a consistent characteristic in PTSD, and thus the relevance of therapies aimed at reducing it to improve symptoms. Although Zhang et al. recognize this problem, they do not address that the reliability of AB measures based on response times, especially with dot-probe tasks, has been questioned. This has led researchers to explore alternative methods for evaluating these biases. 7 Eye-tracking, which infers attention allocation to stimuli from eye movements and fixations during direct exposure has confirmed that individuals with PTSD tend to allocate more attentional resources to aversive stimuli.8,9 In addition, we recently showed that the first moments of exposure to neutral and negative stimuli are characterized by greater engagement of attention toward aversive stimuli in both healthy individuals and individuals with PTSD.10,11 However, this bias is more pronounced among the latter and increases with prolonged exposure to stimuli, unlike healthy individuals who tend to balance their visual exploration over time.
Drawing on these aspects, we conclude that current evidence does not establish the efficacy of ABM or ACT on AB or PTSD-related outcomes. Future studies should favor (i) the inclusion of robust control conditions; (ii) the use of reliable measures like those provided by eye-tracking; (iii) targeting AB components specifically linked to PTSD; and (iv) exploring the relationship between modifications in these components and PTSD symptom alleviation.
