Related Experiment Video
Updated: May 11, 2025
Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments
Published on: January 23, 2017
Measuring attentional bias using the dot-probe task in young women: Psychometric properties and feasibility of
Sandra Klonteig1, Elise S Roalsø2, Brage Kraft3
1Faculty of Technology, Art, and Design, Oslo Metropolitan University, Norway.
Background:
Attentional bias (AB) is characterized by preferential cognitive and emotional processing of mood-congruent stimuli and considered a central mechanism in mood disorders. Considerable research has focused on improving AB measures to enhance mechanistic understanding and clinical utility. The present study examines psychometric properties of a range of AB measures with a multimodal setup.
Methods:
A nonclinical sample of 62 women aged 20-30 years completed the facial dot-probe task while behavioral responses (reaction time), eye-gaze patterns (eye tracking), and electrical brain potentials (electroencephalography) were recorded. AB metrics from four types of AB measures - traditional, response-based, dwell time, and the N2pc component- were examined with internal consistency and short-term test-retest calculations. AB metrics with an internal consistency score over .4 were considered reliable, and their validity were explored by examining relations to depression and anxiety symptoms. In addition, the consistency between reliable metrics across trials were examined.
Results:
Findings show that traditional AB metrics exhibited no degree of reliability, whereas response-based and dwell time metrics overall demonstrated better internal consistencies. Response-based metrics also had higher test-retest reliability in all but one metric. The previously reported reliability of the N2pc component was not observed. As for validity, no linear associations were found between the reliable measures, depression, and anxiety. There were no relations between metrics across trials.
Conclusions:
This study provides insights for future AB research, emphasizing the potential of novel metrics over traditional ones and the use of multimodal setups to develop reliable and potentially hybrid measurements for clinical assessment.
