Thomas Suslow
Dennis Hoepfel
Anette Kersting
Vivien Günther- Department of Psychosomatic Medicine and Psychotherapy, University of Leipzig Medical Center, Leipzig, Germany
Attention orientation toward positive stimuli may have mood-protective or mood-enhancing effects. Eye-tracking is an increasingly administered method to assess biased attention allocation and maintenance. In the present paper, we point to an underestimated but reliable method of eye-tracking research for measuring positive attentional bias and its temporal dynamics: the free-viewing paradigm. To date, few free-viewing eye-tracking studies have specifically examined positive attentional biases in healthy individuals. Against this background, we summarize findings from clinical and subclinical eye-tracking research using free viewing in healthy control groups. We discuss the observed time courses of positive attentional biases during experimental trials, which vary depending on type and number of presented stimuli, and make recommendations on which experimental conditions appear to be favorable for capturing dynamic time courses of positive attentional biases. We identify various individual difference factors that may influence the magnitude of positive attentional biases and should be considered in future studies. Time course analyses of eye-tracking data offer the opportunity to learn more about the time of onset and extent of increase in attention to positive information during free viewing and their relationships to individual difference variables. Directions for future research on positive attentional biases are discussed.
Introduction
From an evolutionary perspective, positive thinking can be seen as a survival mechanism, which helps people to navigate through difficult situations and increase the likelihood of achieving favorable outcomes (1). Healthy individuals experience positive emotions more frequently than negative ones in everyday life (2). According to cross-national data, humans show a general preponderance of positive emotion over negative emotion, with the possible exception of very poor societies (3). Higher positivity ratios are associated with better mental health (4). Various cognitive positivity biases may contribute to this predominance of positive emotions. A positive bias can refer to the tendency to recall positive aspects of past events more vividly than negative ones (5), the tendency to be optimistic about one’s future (6, 7), or self-serving attributional tendencies, making more internal, stable, and global attributions for positive than for negative events (8). Another form of positivity bias is observed in the area of attention processes: individuals appear to prioritize and pay more attention to positive or rewarding stimuli compared to neutral ones (9). A positive attentional bias may have beneficial effects on mood, stress resilience, and social contact (10–12).
An important feature of attention is its selectivity, i.e., cognitive resources are focused on certain aspects of the environment rather than on others (13). Selective processes can be observed at early and late stages of perception. In the area of emotion perception, automatic vigilance for negative and especially threatening stimuli has been documented during early information processing (14, 15). Clinical research on attentional biases has shown that individuals with anxiety disorders exhibit a particularly facilitated detection and orienting to threat-related stimuli compared to healthy individuals (16). In the last decades, clinical studies of attentional biases have frequently relied on reaction time tasks such as the dot-probe task (17), the emotional Stroop task (18), or the face-in-the-crowd task (19). These tasks are indirect measures of attention, which require a motor response and depend on speed-accuracy criteria. They rely on reactions to stimuli measured at a specific point in time. A serious methodological problem of reaction-time measures lies in the poor reliability of the attention parameters derived (20–22).
Recently, eye-tracking technology has gained increasing popularity in various research areas (23, 24). Eye-tracking allows continuous recording of eye movements and represents a direct method to assess attention allocation, as direction of eye gaze and focus of attention are often closely linked (25). The highest visual quality is achieved within a small foveal region at the center of gaze that covers about 2° of visual angle (26). Where the eyes look determines which visual information is focused on and perceived in more detail. Eye-tracking has become one of the most important experimental methods in the study of attentional biases in mental disorders (27–31). It can be combined with various experimental paradigms such as the dot probe (32) or the face-in-the-crowd task (33), e.g., to assess early and late attention allocation or scan paths. It is interesting to note that when eye movements are registered and analyzed during dot probe tasks, eye-tracking and traditional reaction-time based attention parameters are largely unrelated (34–36). It appears that these different attention bias measures could reflect different stages of processing or attention types (overt vs. covert) (37, 38). Therefore, registration of eye-movements during the dot probe task does not appear to be a kind of substitute for reaction times in determining attention allocation. However, eye-tracking indices of late attentional processes such as bias scores based on dwell time or fixation frequency tend to yield higher reliability coefficients (internal consistency) than traditional reaction-time based bias scores (34, 36). In the dot probe task, eye-tracking indices of early attentional processes, such as the proportion of first fixations or the first fixation latency bias, also show poor reliability (34).
A frequently used paradigm in eye-tracking research to measure attentional biases in emotion perception is free-viewing. In the next section, we will introduce the free-viewing paradigm and discuss methodological features such as task instructions, type and number of stimuli used and the reliability problem of the attention parameters derived from the task. In our paper, we then present the evidence for a positive attentional bias in healthy individuals as reported in clinical studies, since there is little specific research on positive attentional bias in healthy individuals. The literature discussed comes primarily from recent meta-analyses on attention bias in depression and anxiety (28–30, 39) and additional searches in the databases Web of Science and PsychINFO. Afterwards, we point out various individual difference factors that can influence the magnitude of positive attentional biases during free viewing. Finally, we deal with the time course of visual attention during free viewing, since little is known about the onset of positive attentional biases but fast gaze orientation toward positive stimuli could be important for efficient mood stabilization and repair.
The free-viewing paradigm
The free-viewing paradigm, sometimes also called passive viewing paradigm, refers to a research approach where participants are asked to observe one or more stimuli (such as images, words, or videos) without any specific task. This setup enables the investigation of spontaneous gaze behavior and attention allocation. Free-viewing tasks are relatively simple to implement, making them suitable for a wide range of research questions and study populations including, for example, infants, apes, and dogs (40–42).
In a laboratory setting, participants’ only task is to observe the stimuli shown on the monitor. This research methodology contrasts with traditional experimental paradigms where participants are instructed to process visual stimuli in a specific way (e.g., judge the age of individuals shown) while viewing the stimuli. Free-viewing tasks differ from traditional reaction time tasks assessing attention processes, where, for example, it is necessary to respond as quickly as possible to the location of a dot that replaces one of two stimuli (as in dot probe tasks (17)), to determine the color of a word as quickly as possible, while ignoring the word’s meaning (as in emotional Stroop tasks (18)) or to look at a series of faces and search for a discrepant emotional expression as quickly as possible (as in face-in-the-crowd tasks (19)).
The exact instructions in free-viewing tasks can vary across studies. For example, in some investigations, participants were asked to look at the images in any fashion they wished (43), to view the images naturally, with no further requirements (44), or to look attentively at the stimuli presented (45). Future research has to clarify whether instructions that encourage more attentive viewing could elicit more exploratory, information-seeking gaze behavior compared to instructions that emphasize freedom and spontaneity of viewing. Attentive viewing instructions may increase exploratory behavior, depth of stimulus encoding, and accelerate eye movements compared to more passive viewing instructions (46).
In free-viewing tasks, either a single stimulus (e.g., face or image) can be presented or multiple stimuli (e.g., 2, 4, or 16) can be displayed simultaneously. When only one stimulus is shown, it can be examined which details of the stimulus are viewed first and for how long (47, 48). The presentation of multiple stimuli offers the opportunity to investigate processes of early and late preference for one stimulus category over another. Early attention allocation can be assessed through parameters such as location (or probability) of first fixation or first fixation latency, i.e., the time it takes to fixate on a stimulus. Parameters like dwell time or fixation duration measure processes of sustained attention allocation and refer to the length of time the gaze remains focused on a specific stimulus. A serious methodological problem of early attention parameters derived from multiple-stimulus free-viewing tasks lies in their poor reliability, i.e., low internal consistency, split-half and test-retest reliability (34, 49, 50). An important factor contributing to these poor reliability results for early parameters could be culturally shaped initial gaze behavior, which is directed to the left in Western and to the right in Eastern cultures, when one should read or look at something. In contrast, dwell time and fixation duration measures derived from multiple stimulus free-viewing tasks yield in general adequate to good internal consistencies and split-half reliabilities (34, 49–51) and, in many cases, acceptable test-retest reliabilities (49, 50, 52). Reliable, psychometrically sound attention parameters are necessary for promoting our understanding of the attention processes implicated in psychopathology (53).
Researchers wanting to use free-viewing tasks should ensure that study-relevant contents or stimulus categories are not primed by prior presentation of another experiment or prior completion of questionnaires. Primed features can automatically guide attention to stimuli with the same features as the prime in subsequent displays, no matter whether prime features were implicitly or explicitly encoded (54).
Positive attentional bias as assessed during free viewing: the effect of depression, mood states, individual differences, and age
Free viewing has been widely used in clinical research to investigate decrease in positive attentional bias in patients with depressive disorders (30, 39). In addition to an increased attention for dysphoric information, clinically depressed patients are characterized by reductions in attention maintenance for positive stimuli compared to healthy individuals. The two forms of attentional distortions in depression, i.e., more sustained attention to dysphoric stimuli and reduced attention allocation to positive stimuli, can occur independently of each other. These attentional biases have been demonstrated separately in experiments in which a sad (or happy) face was presented alongside a neutral face (44, 55). The studies in the field focused on group differences between depressed and non-depressed individuals and hardly addressed questions concerning the magnitude of attentional preferences for positive over negative information in the non-depressed control groups. An inspection of the data from the relevant free-viewing studies provides evidence that the gaze of healthy individuals (at least descriptively) remains longer on positive than on negative stimuli - regardless of whether faces (shown for 3 to 10 seconds) or images (shown for 30 seconds) were presented in the experiments (faces (44, 51, 56–59), images (60, 61) – but see (55, 62) for discrepant results). The two studies with discrepant results were both face-based and presented only two faces. Moreover, the proportion of women in these investigations was slightly lower (53% and 56%) than in the other studies (61-75%) (44, 56–59). Thus, the detection of a positive attentional bias may be more difficult when pairs of faces are presented than when four or more faces are used (56–58). A low proportion of women in a sample could also make it more difficult to detect positive bias effects. Decreases in sustained attention for positive information during free viewing have also been observed in dysphoric or mildly depressed compared to non-depressed individuals (63, 64).
In research on biases in sustained attention allocation or attentional preferences, indices were used that are based on dwell time, fixation duration or number of fixations. In contrast to fixation duration, dwell time includes also the time spent on a specific area of interest during gaze movements. In many studies, the analysis for determining attentional biases was based on the mean dwell or fixation time (in seconds or milliseconds) to emotional and neutral stimuli, which were then entered into statistical analyses (57, 58, 61, 63–65). In several other studies, relative bias scores were calculated, for example, by subtracting the fixation time to neutral faces from that to a specific emotion face category (44, 55), dividing the fixation time to a specific emotion face by the fixation time spent on a multiple-stimulus array (59), calculating the percentage of time on a specific emotion category compared to all faces (60) or dividing the fixation time to a specific emotion face by the fixation time spent on the specific emotion face and the paired neutral face (34). Although raw dwell time or fixation duration values are often used in eye-tracking based attention bias research, it seems advisable to compute relative bias scores in the case of studies that examine attention to different emotions and present only two stimuli per trial (an emotional combined with a neutral stimulus). Bias scores that consider the total fixation time for both stimuli may improve here the comparability of the magnitude of attention allocation between emotion categories.
Mood states can influence attention allocation in non-depressed individuals during free viewing. Increases in happy mood after positive mood induction was found to be linked to enhanced attentional deployment to happy faces (66). Increased negative mood (in response to negative mood induction) is associated with increased attentional deployment to positive images (43, 67). The latter finding suggests that heightened attention to positive stimuli could be part of an emotion regulation strategy related to mood repair.
Various other individual difference factors can influence the magnitude of positive attentional biases during free viewing. Trait happiness and life satisfaction were found to be positively associated with attention maintenance on positive compared to neutral scenes, irrespective of category, i.e. achievement, social, and primary reward (68). These effects were especially prominent during the later phases of sustained viewing. The personality trait of alexithymia, which is characterized by difficulties identifying and expressing one’s emotions, could attenuate positive attentional biases. Surber et al. (69) observed in non-alexithymic individuals an attentional preference for happy over angry and neutral facial expressions, whereas in alexithymic individuals no attentional preferences were determined. Experiences of maltreatment during childhood may affect the perception of positive content in adulthood. Attentional preference for positive over other emotional faces during free viewing was found to be diminished in women with experiences of physical and emotional abuse in childhood (70). Reductions in attentional preference for positive stimuli may be one pathway through which experiences of severe abuse increase the risk for the development of depressive symptoms and affective disorders.
There are indications that age might be associated with heightened orientation toward positive stimuli during free viewing (71). According to Faul et al. (72) the emergence of the positivity effect in older adults could be related to absence of depressive symptoms and reappraisal regulatory preferences. Thus, in healthy older people preference for using reappraisal in daily life could be related to an increased positive attentional bias.
Time course of visual attention to positive information
There are eye-tracking studies based on free-viewing tasks that have examined changes in attention allocation not only as a function of emotional quality of stimuli, but also of time. Some of these studies provide information on the time course of attention to positive stimuli compared to other simultaneously shown emotional or neutral stimuli in healthy individuals. Kellough et al. (60) presented happy, sad, threatening and neutral images for 30 seconds to depressed and never depressed individuals. The never depressed individuals were young university students (with an age range between 18 and 21 years; 47% were women). Stable differences between study groups in fixation time were observed for dysphoric and positive faces over time. In the total sample, there was a strong initial attention to threat images during the first 5 seconds of presentation that decreased substantially during the next 5 seconds. In contrast, attention to positive images was initially less pronounced but tended to increase during the experiment. Arndt et al. (64) also used positive, dysphoric, threatening, and neutral images showing them simultaneously for 10 seconds to dysphoric and non-dysphoric individuals. The non-dysphoric individuals were university students (with a mean age of 21 years; only women were included). In the non-dysphoric group, during the first two seconds the longest fixation times were observed for threat images, which decreased subsequently. In contrast, the positive images were viewed less during the first four seconds compared with the threat images, but thereafter there was a steep increase in fixation time for positive images. After eight to ten seconds, positive images were looked at much longer than the other image categories.
In the study of Waechter et al. (34), two faces from the same model were presented simultaneously (a happy, angry, or disgusted expression combined with a neutral one) for 5 seconds to individuals with high and low social anxiety. Social anxiety had no effect on attention but an emotion by time interaction was observed. Study participants were university students (with a mean age of 19 years; 67% were women). Proportion of viewing time was analyzed in 500ms intervals. When viewing anger-neutral face pairs, participants showed a significant attention bias toward anger faces from 501 to 2.500ms. When looking at disgust-neutral pairs, a bias toward disgust faces was found only between 501 and 1.500ms. In the last four 500ms intervals participants looked longer at the neutral compared to the angry or disgusted face (at a descriptive level). When viewing happy-neutral pairs, an attentional bias toward happy faces was observed from the second interval (i.e., 501-1.000ms) to the last interval (i.e., 4.501-5.000ms). In the study of Byrow et al. (65), socially anxious and non-anxious individuals were presented two faces of the same model (a happy or angry expression combined with a neutral one) for 1.5 seconds. The non-anxious individuals were university students (with a mean age of 26 years; 48% were women). Healthy controls showed more fixations on emotional than on neutral faces. No differences in percentage of fixations were observed between happy and angry faces. Fernandes et al. (73) compared visual attention of high- and low-socially anxious individuals showing two faces of the same model (combinations of happy, angry, and neutral expressions) for 1.5 seconds. The low anxious individuals were university students (with a mean age of 22 years; 63% were women). In the low anxiety group, there were no differences in dwell time between angry and happy faces, but dwell time decreased for both emotional qualities over the 1.5 seconds. Gamble and Rapee (74) presented two faces from the same model (a happy or angry expression paired with a neutral one) for 5 seconds to social phobic and non-phobic individuals. 43% of the non-phobic individuals were university students and 57% were community volunteers (with a mean age of 36 years; 43% were women). In the healthy control group, attention for happy faces (compared to neutral faces) was higher compared to attention for angry faces throughout the five 1-second intervals of the experiment. Soltani et al. (75) presented sets consisting of happy, sad, threatening, and neutral faces of different models for 8 seconds to depressed and never depressed individuals. The never depressed individuals were university students or community members (with a mean age of 24 years; only women were included). Already in the first 2-second interval, the never depressed subjects looked at the happy faces the longest. In the never depressed individuals, attention to happy faces increased continuously over the four intervals.
To summarize, the results of the studies based on (four) images indicate an initial attention allocation to threat stimuli (60, 64). Both studies are based on samples of young university students. In the study with a long trial duration, i.e., 30s (60), a longer initial attentional prioritization of threatening images was found compared to the study with a short trial duration, i.e., 10s (64). Possibly, the total period of time available to the participants for observation could play a role in the temporal development of attention allocation. However, across both studies (60, 64) there is evidence that positive images are initially less attended but subsequently they receive more attention. The findings of the face-based investigations show a different picture of attention allocation over time. When four faces are simultaneously presented positive faces are viewed the longest already in the first two seconds followed by a further increase in attention (75). As in the image-based investigation of Arndt et al. (64) study participants of Soltani et al. (75) were young, female and at least in part university students so that differences in the findings on the time course of attention should not be due to the variables age or biological sex. It is interesting to note that when one angry or disgusted face is presented together with a neutral one the initial attention bias toward the threatening (or hostile) face is no greater than toward a happy face (34). Subsequently, anger and disgust faces can lose attention, whereas happy faces may continue to receive more attention than neutral faces over time. There is even evidence that happy faces (in pairs with neutral faces) receive more attention than angry faces in the first second of presentation (74). This early prioritization of positive faces was observed in a sample of individuals who were on average in their mid-thirties and thus somewhat older than the participants in the other studies (34, 60, 64, 65, 73, 75). This raises the question of whether, with increasing age during young adulthood, the attentional prioritization of positive faces might start earlier in the perception process. None of the above cited free-viewing eye-tracking studies on attentional biases administered dynamic stimuli (dynamic facial expressions or video clips).
Discussion: conclusions and directions for future research
Considering the above findings, the free-viewing paradigm appears generally well suited to capture positive attentional biases in healthy individuals. However, the number, type and duration of the presented stimuli appear to have a substantial influence on the occurrence of the bias. Our analysis of the literature shows that, to date, little research has been conducted that focused on the analysis of the temporal development of positive attentional biases in non-patients. The existing studies were based on samples consisting entirely or partly of university students. This limits the generalizability of the results on the time-course of positive attentional biases and indicates a need for future research to also examine less educated and older individuals. Attentional preferences for positive environmental stimuli could be important determinants of positive affect (10, 12), which should be better understood in terms of their conditions of occurrence. Attention orientation toward positive stimuli could serve as a relatively automatic, low-effort regulatory strategy that promotes mood (68). Directing and holding one’s attention on positive facial expression could act as one mechanism, by which positive affect is elicited or enhanced in everyday life. Processes of emotional contagion might play a crucial role in inducing positive affective reactions during the perception of happy facial expression. Emotional contagion describes the phenomenon that people automatically mimic and synchronize their emotions with those of another person (76) and is assumed to occur through facial mimicry and feedback (77). Receivers imitate senders’ emotional displays in emotional mimicry, even if these are only of short duration (78). Facial feedback from such mimicry processes can activate the corresponding emotional state in receivers. The perception of happy facial expressions can elicit positive feelings in the receivers through facial muscle activity (79).
Importantly, internal consistency and split-half reliability of positive attention bias scores (dwell times) for single time intervals were found to be adequate to good for images as well as faces - except for the first two seconds (50). Individual differences in the propensity to look at positive stimuli are rather stable across contents and across a two-months period (80). In previous eye-tracking studies, reliability data on attention bias parameters were rarely provided, but this should always be done in future studies in order to be able to assess the solidity of the reported findings.
When selecting the stimuli for free-viewing tasks, it seems important to consider that images (i.e., scenes) require longer trials. Compared to faces (from the same model), scenes are more complex visual stimuli with a much wider range of objects and stimulus constellations shown. Scenes require a longer period of time to process and recognize their contents. This could be a reason why there is a longer initial attention allocation to threatening scenes during the first seconds, which has not been found for angry faces. Efficiency of processing facial expressions of basic emotions is illustrated by the fact that the quality of briefly presented emotional faces can be processed even if participants are not aware of their presentation (81, 82). If one is interested in investigating temporal changes in positive attentional bias over a trial, the presentation of four stimuli seems more suitable than that of two stimuli (a positive stimulus combined with a neutral one) and (naturalistic) images seem more suitable than faces. It can be assumed that in multiple-stimulus presentations positive faces attract the attention of viewers much faster than positive images (scenes). The relatively late prioritization of positive images should offer better opportunities to capture the temporal dynamics of the attention allocation processes and interindividual differences in the time course. To assess positive attentional biases relevant for mood protection, it may be important to administer arrays in which positive stimuli are presented along with negative stimuli (and not only with neutral ones). The presence of negative facial expressions in the perceptual field may trigger or reinforce the need for mood regulation. Possibly, the orientation toward positive information in the face of negative information represents an important mood-protective attentional mechanism. It remains to be examined whether the magnitude of a positive bias measured by simultaneously presenting positive and negative stimuli has a greater predictive value with regard to positive mood and depressive symptoms than a positive bias measured by presenting positive stimuli combined with neutral stimuli.
The free-viewing task appears to be particularly suited to capturing interindividual differences in mood-congruent effects on attention. Memory or evaluation tasks that require participants to purposefully attend to each image, regardless of image type, could lead to a more uniform distribution of attention across stimuli that might attenuate individual differences in attention allocation (83). Recently, Sun et al. (84), used simultaneous eye-tracking and fMRI measurements to assess emotional attentional biases in depressed and non-depressed individuals. Gaze behavior during emotion recognition was compared with gaze behavior during free viewing based on the same stimuli. A mood congruent pattern was found in depressed patients only in the free-viewing condition. Emotion recognition was linked to greater activation of the primary visual cortex, whereas free viewing was more strongly associated with activity in the dorsolateral prefrontal cortex. Emotion identification may lead to a more feature-based visual processing while free viewing involves more spontaneous, self-generated attentional responses depending on individual state and trait characteristics.
Individual factors that can influence the magnitude of positive attentional biases and should be considered in future research are trait happiness, alexithymia, experiences of abuse during childhood, habitual use of reappraisal (as emotion regulation strategy), and age. With increasing age, positive stimuli seem to capture attention more quickly and to hold it for longer. It should be clarified whether early onset or the extent of increase in attention to positive information during free viewing are better predictors of positive mood or absence of depressive symptoms than positive bias scores averaged over the task. Future research should investigate whether positive attentional biases are related to adaptive strategies of emotion regulation such as acceptance and problem-solving (85), and other cognitive positivity biases (e.g., in memory processes (5)) and how they interact to protect or improve mood. In this research context, mediation analyses could be a valuable tool to investigate whether preference for positive stimuli forms an attentional pathway through which adaptive emotion regulation strategies (such as cognitive reappraisal or acceptance) promote positive mood and buffer the development of negative affect (86).
Clinical studies investigating positive attentional biases have so far focused on depressive and (social) anxiety disorders. Eye-tracking research into positive attentional biases in other mental health conditions, such as schizophrenia and post-traumatic stress disorder, is less developed. Therefore, future research efforts using the free-viewing task can be directed, for example, towards clarifying whether anhedonia symptoms in schizophrenia or impaired emotion regulation in PTSD are associated with reduced spontaneous attention allocation to positive stimuli.
A novel method that was developed as an alternative to eye tracking is MouseView.js, which uses the computer cursor as a proxy for gaze instead of a webcam (87). This program mimics the visual system’s foveal clarity and peripheral blur by allowing participants to move an aperture on an obscured field with the mouse. There is evidence that MouseView.js approximates results from eye tracking when participants engage in visual exploration of stimuli during free-viewing tasks (87). MouseView.js is a reliable measure of attentional preferences and can replace eye tracking, especially in web-based psychological experiments. Recent studies on the effects of anxiety disorders and sexual orientation on perception demonstrated the utility of MouseView.js for capturing attentional biases (88, 89).
So far, free-viewing eye-tracking studies on attention biases using multiple-stimulus arrays have almost exclusively used static stimuli, i.e., images of faces or images of scenes or objects. Static images cannot adequately depict the dynamic nature of real-life situations and interactions. The use of dynamic stimuli (videos) could strengthen research on attention biases with regard to ecological validity of findings. In a recent eye-tracking study on attention bias in young adults (90), positive, negative, and neutral video clips depicting social situations were shown in side-by-side pairs. Individuals with greater anxiety were found to spend more time gazing at negative compared to paired neutral videos. However, even if videos generally seem promising for attention research, it must be ensured that the amount and speed of motion in the video clips are comparable across different experimental conditions. This represents an important methodological challenge for future research on attentional biases in which videos instead of images are administered.
Future research into positive attentional biases should not only rely on faces or pictures as stimuli but could also include news websites that consist of positive and negative articles. Rudich-Strassler et al. (91) emphasize the importance of ecological validity in attentional research which can be increased through the use of internet websites that resemble real-world online environments.
Data availability statement
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.
Author contributions
TS: Conceptualization, Writing – original draft. DH: Writing – review & editing. TW: Writing – review & editing. AK: Writing – review & editing. VG: Conceptualization, Writing – review & editing.
Funding
The author(s) declare that financial support was received for the research and/or publication of this article. This work was supported by grants from the German Research Foundation DFG to VG (GU 2231/2-1, project number: 496949003).
Acknowledgments
The authors acknowledge that this publication was funded by the Open Access Publishing Fund of Leipzig University supported by the German Research Foundation within the program Open Access Publication Funding.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declare that no Generative AI was used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1. Shiota MN. The evolutionary perspective in positive emotion research. In: Tugade MM, Shiota MN, and Kirby LD, editors. Handbook of Positive Emotions. Guilford Press, New York (2014). p. 44–59.
2. Watson D and Walker LM. The long-term stability and predictive validity of trait measures of affect. J Pers Soc Psychol. (1996) 70:567–77. doi: 10.1037/0022-3514.70.3.567
3. Diener E and Diener C. Most people are happy. Psychol Sci. (1996) 7:181–5. doi: 10.1111/j.1467-9280.1996.tb00354.x
4. Diehl M, Hay EL, and Berg KM. The ratio between positive and negative affect and flourishing mental health across adulthood. Aging Ment Health. (2011) 15:882–93. doi: 10.1080/13607863.2011.569488
5. Adler O and Pansky A. A “rosy view” of the past: Positive memory biases. In: Aue T and Okon-Singer H, editors. Cognitive biases in health and psychiatric disorders. Academic Press, San Diego (2020). p. 139–71.
6. Hoorens V, Smits T, and Shepperd J. Comparative optimism in the spontaneous generation of future life events. Br J Soc Psychol. (2008) 47:441–51. doi: 10.1348/014466607X236023
7. Lench HC and Bench SW. Automatic optimism: Why people assume their futures will be bright. Soc Pers Psychol Compass. (2012) 6:347–60. doi: 10.1111/j.1751-9004.2012.00430.x
8. Mezulis AH, Abramson LY, Hyde JS, and Hankin BL. Is there a universal positivity bias in attribution? A meta-analytic review of individual, developmental, and cultural differences in the self-serving attributional bias. Psychol Bull. (2004) 130:711–47. doi: 10.1037/0033-2909.130.5.711
9. Pool E, Brosch T, Delplanque S, and Sander D. Attentional bias for positive emotional stimuli: A meta-analytic investigation. Psychol Bull. (2016) 142:79–106. doi: 10.1037/bul0000026
10. Thoern HA, Grueschow M, Ehlert U, Ruff CC, and Kleim B. Attentional bias towards positive emotion predicts stress resilience. PLoS One. (2016) 1:e0148368. doi: 10.1371/journal.pone.0148368
11. Troller-Renfree S, McLaughlin KA, Sheridan MA, Nelson CA, Zeanah CH, and Fox NA. The beneficial effects of a positive attention bias amongst children with a history of psychosocial deprivation. Biol Psychol. (2017) 122:110–20. doi: 10.1016/j.biopsycho.2016.04.008
12. Kress L and Aue T. Learning to look at the bright side of life: Attention bias modification training enhances optimism bias. Front Hum Neurosci. (2019) 13:222. doi: 10.3389/fnhum.2019.00222
13. Proctor RW and Vu KPL. Attention: Selection and control in human information processing. Washington: American Psychological Association (2023).
14. Estes Z and Adelman JS. Automatic vigilance for negative words is categorical and general. Emotion. (2008) 8:453–57. doi: 10.1037/a0012887
15. Anderson BA and Britton MK. On the automaticity of attentional orienting to threatening stimuli. Emotion. (2020) 20:1109–112. doi: 10.1037/emo0000596
16. Pergamin-Hight L, Naim R, Bakermans-Kranenburg MJ, van IJzendoorn MH, and Bar-Haim Y. Content specificity of attention bias to threat in anxiety disorders: a meta-analysis. Clin Psychol Rev. (2015) 35:10–8. doi: 10.1016/j.cpr.2014.10.005
17. MacLeod C, Mathews A, and Tata P. Attentional bias in emotional disorders. J Abnorm Psychol. (1986) 95:15–20. doi: 10.1037/0021-843X.95.1.15
18. Williams JM, Mathews A, and MacLeod C. The emotional Stroop task and psychopathology. Psychol Bull. (1996) 120:3–24. doi: 10.1037/0033-2909.120.1.3
19. Hansen CH and Hansen RD. Finding the face in the crowd: An anger superiority effect. J Pers Soc Psychol. (1988) 54:917–24. doi: 10.1037/0022-3514.54.6.917
20. Strauss GP, Allen DN, Jorgensen ML, and Cramer SL. Test-retest reliability of standard and emotional stroop tasks: an investigation of color-word and picture-word versions. Assessment. (2005) 12:330–37. doi: 10.1177/1073191105276375
21. Schmukle SC. Unreliability of the dot probe task. Eur J Pers. (2005) 19:595–605. doi: 10.1002/per.554
22. Staugaard SR. Reliability of two versions of the dot-probe task using photographic faces. Psychol Sci Q. (2009) 51:339–50.
23. Rahal RM and Fiedler S. Understanding cognitive and affective mechanisms in social psychology through eye-tracking. J Exp Soc Psychol. (2019) 85:103842. doi: 10.1016/j.jesp.2019.103842
24. Tahri Sqalli M, Aslonov B, Gafurov M, Mukhammadiev N, and Sqalli Houssaini Y. Eye tracking technology in medical practice: a perspective on its diverse applications. Front Med Technol. (2023) 5:1253001. doi: 10.3389/fmedt.2023.1253001
26. Helo A, Pannasch S, Sirri L, and Rämä P. The maturation of eye movement behavior: scene viewing characteristics in children and adults. Vision Res. (2014) 103:83–91. doi: 10.1016/j.visres.2014.08.006
27. Hagan KE, Alasmar A, Exum A, Chinn B, and Forbush KT. A systematic review and meta-analysis of attentional bias toward food in individuals with overweight and obesity. Appetite. (2020) 151:104710. doi: 10.1016/j.appet.2020.104710
28. Günther V, Kropidlowski A, Schmidt FM, Koelkebeck K, Kersting A, and Suslow T. Attentional processes during emotional face perception in social anxiety disorder: A systematic review and meta-analysis of eye-tracking findings. Prog Neuropsychopharmacol Biol Psychiatry. (2021) 111:110353. doi: 10.1016/j.pnpbp.2021.110353
29. Clauss K, Gorday JY, and Bardeen JR. Eye tracking evidence of threat-related attentional bias in anxiety- and fear-related disorders: A systematic review and meta-analysis. Clin Psychol Rev. (2022) 93:102142. doi: 10.1016/j.cpr.2022.102142
30. Huang G, Li Y, Zhu H, Feng H, Shen X, and Chen Z. Emotional stimulation processing characteristics in depression: Meta-analysis of eye tracking findings. Front Psychol. (2023) 13:1089654. doi: 10.3389/fpsyg.2022.1089654
31. Hinz JR, Eikeseth FF, Chawarska K, and Eikeseth S. A systematic review and meta-analysis of atypical visual attention towards non-social stimuli in preschoolers with autism spectrum disorder. Autism Res. (2024) 17:2628–44. doi: 10.1002/aur.3261
32. Meregalli V, Giovannini S, Trevisan A, Romanelli M, Ugur S, Tenconi E, et al. Eyes on the body: Assessing attentional bias toward body-related stimuli in Anorexia Nervosa. J Psychiatr Res. (2025) 182:506–12. doi: 10.1016/j.jpsychires.2025.01.043
33. Blekic W, Rossignol M, and D’Hondt F. Examining attentional avoidance in post-traumatic stress disorder: an exploratory ‘Face in the Crowd’ paradigm using eye-tracking. Eur J Psychotraumatol. (2025) 16:2462489. doi: 10.1080/20008066.2025.2462489
34. Waechter S, Nelson AL, Wright C, Hyatt A, and Oakman J. Measuring attentional bias to threat: Reliability of dot probe and eye movement indices. Cognit Ther Res. (2014) 38:313–33. doi: 10.1007/s10608-013-9588-2
35. Hoepfel D, Günther V, Bujanow A, Kersting A, Bodenschatz CM, and Suslow T. Experiences of maltreatment in childhood and attention to facial emotions in healthy young women. Sci Rep. (2022) 12:4317. doi: 10.1038/s41598-022-08290-1
36. Klonteig S, Roalsø ES, Kraft B, Moberget T, Hilland E, Mirtaheri P, et al. Measuring attentional bias using the dot-probe task in young women: Psychometric properties and feasibility of response-based computations, dwell time, and the N2pc component. J Behav Ther Exp Psychiatry. (2025) 88:102036. doi: 10.1016/j.jbtep.2025.102036
37. Petrova K, Wentura D, and Bermeitinger C. What happens during the stimulus onset asynchrony in the dot-probe task? Exploring the role of eye movements in the assessment of attentional biases. PloS One. (2013) 8:e76335. doi: 10.1371/journal.pone.0076335
38. Pasqualette L and Kulke L. Differences between overt, covert and natural attention shifts to emotional faces. Neuroscience. (2024) 559:283–92. doi: 10.1016/j.neuroscience.2024.09.009
39. Suslow T, Hußlack A, Kersting A, and Bodenschatz CM. Attentional biases to emotional information in clinical depression: A systematic and meta-analytic review of eye tracking findings. J Affect Disord. (2020) 274:632–42. doi: 10.1016/j.jad.2020.05.140
40. Kis A, Hernádi A, Miklósi B, Kanizsár O, and Topál J. The way dogs (Canis familiaris) look at human emotional faces is modulated by oxytocin. An eye-tracking study. Front Behav Neurosci. (2017) 11:210. doi: 10.3389/fnbeh.2017.00210
41. Portugal AM, Viktorsson C, Taylor MJ, Mason L, Tammimies K, Ronald A, et al. Infants’ looking preferences for social versus non-social objects reflect genetic variation. Nat Hum Behav. (2024) 8:115–24. doi: 10.1038/s41562-023-01764-w
42. van Berlo E, Roth TS, Kim Y, and Kret ME. Selective and prolonged attention to emotional scenes in humans and bonobos. Proc Biol Sci. (2024) 291:20240433. doi: 10.1098/rspb.2024.0433
43. Newman KR and Sears CR. Eye gaze tracking reveals different effects of a sad mood induction on the attention of previously depressed and never depressed women. Cognit Ther Res. (2015) 39:292–306. doi: 10.1007/s10608-014-9669-x
44. Duque A and Vázquez C. Double attention bias for positive and negative emotional faces in clinical depression: evidence from an eye-tracking study. J Behav Ther Exp Psychiatry. (2015) 46:107–14. doi: 10.1016/j.jbtep.2014.09.005
45. Schomaker J and Wittmann BC. Memory performance for everyday motivational and neutral objects is dissociable from attention. Front Behav Neurosci. (2017) 11:121. doi: 10.3389/fnbeh.2017.00121
46. Haskins AJ, Mentch J, Botch TL, and Robertson CE. Active vision in immersive, 360° real-world environments. Sci Rep. (2020) 10:14304. doi: 10.1038/s41598-020-71125-4
47. Thompson SJ, Foulsham T, Leekam SR, and Jones CRG. Attention to the face is characterised by a difficult to inhibit first fixation to the eyes. Acta Psychol. (2019) 193:229–38. doi: 10.1016/j.actpsy.2019.01.006
48. Viktorsson C, Portugal AM, Li D, Rudling M, Siqueiros Sanchez M, Tammimies K, et al. Preferential looking to eyes versus mouth in early infancy: heritability and link to concurrent and later development. J Child Psychol Psychiatry. (2023) 64:311–9. doi: 10.1111/jcpp.13724
49. Skinner IW, Hübscher M, Moseley GL, Lee H, Wand BM, Traeger AC, et al. The reliability of eyetracking to assess attentional bias to threatening words in healthy individuals. Behav Res Methods. (2018) 50:1778–92. doi: 10.3758/s13428-017-0946-y
50. Sears C, Quigley L, Fernandez A, Newman K, and Dobson K. The reliability of attentional biases for emotional images measured using a free-viewing eye-tracking paradigm. Behav Res Methods. (2019) 51:2748–60. doi: 10.3758/s13428-018-1147-z
51. Klawohn J, Bruchnak A, Burani K, Meyer A, Lazarov A, Bar-Haim Y, et al. Aberrant attentional bias to sad faces in depression and the role of stressful life events: Evidence from an eye-tracking paradigm. Behav Res Ther. (2020) 135:103762. doi: 10.1016/j.brat.2020.103762
52. Shamai-Leshem D, Abend R, Arad G, Azriel O, Chong L, de Jong P, et al. The free-viewing matrix task: A reliable measure of attention allocation in psychopathology. J Anxiety Disord. (2023) 100:102789. doi: 10.1016/j.janxdis.2023.102789
53. Rodebaugh TL, Scullin RB, Langer JK, Dixon DJ, Huppert JD, Bernstein A, et al. Unreliability as a threat to understanding psychopathology: The cautionary tale of attentional bias. J Abnorm Psychol. (2016) 125:840–51. doi: 10.1037/abn0000184
54. Li A, Wolfe JM, and Chen Z. Implicitly and explicitly encoded features can guide attention in free viewing. J Vis. (2020) 20:8. doi: 10.1167/jov.20.6.8
55. Lu S, Xu J, Li M, Xue J, Lu X, Feng L, et al. Attentional bias scores in patients with depression and effects of age: a controlled, eye-tracking study. J Int Med Res. (2017) 45:1518–27. doi: 10.1177/0300060517708920
56. Isaac L, Vrijsen JN, Rinck M, Speckens A, and Becker ES. Shorter gaze duration for happy faces in current but not remitted depression: evidence from eye movements. Psychiatry Res. (2014) 218:79–86. doi: 10.1016/j.psychres.2014.04.002
57. Lazarov A, Ben-Zion Z, Shamai D, Pine DS, and Bar-Haim Y. Free viewing of sad and happy faces in depression: A potential target for attention bias modification. J Affect Disord. (2018) 238:94–100. doi: 10.1016/j.jad.2018.05.047
58. Bodenschatz CM, Skopinceva M, Ruß T, and Suslow T. Attentional bias and childhood maltreatment in clinical depression - An eye-tracking study. J Psychiatr Res. (2019) 112:83–8. doi: 10.1016/j.jpsychires.2019.02.025
59. Unruh KE, Bodfish JW, and Gotham KO. Adults with autism and adults with depression show similar attentional biases to social-affective images. J Autism Dev Disord. (2020) 50:2336–47. doi: 10.1007/s10803-018-3627-5
60. Kellough JL, Beevers CG, Ellis AJ, and Wells TT. Time course of selective attention in clinically depressed young adults: An eye tracking study. Behav Res Ther. (2008) 46:1238–43. doi: 10.1016/j.brat.2008.07.004
61. Wells TT, Clerkin EM, Ellis AJ, and Beevers CG. Effect of antidepressant medication use on emotional information processing in major depression. Am J Psychiatry. (2014) 171:195–200. doi: 10.1176/appi.ajp.2013.12091243
62. Lagattuta KH and Kramer HJ. Try to look on the bright side: Children and adults can (sometimes) override their tendency to prioritize negative faces. J Exp Psychol Gen. (2017) 146:89–101. doi: 10.1037/xge0000247
63. Ellis AJ, Beevers CG, and Wells TT. Attention allocation and incidental recognition of emotional information in dysphoria. Cognit Ther Res. (2011) 35:425–33. doi: 10.1007/s10608-010-9305-3
64. Arndt JE, Newman KR, and Sears CR. An eye tracking study of the time course of attention to positive and negative images in dysphoric and nondysphoric individuals. J Exp Psychopathol. (2014) 5:399–413. doi: 10.5127/jep.035813
65. Byrow Y, Chen NT, and Peters L. Time course of attention in socially anxious individuals: investigating the effects of adult attachment style. Behav Ther. (2016) 47:560–71. doi: 10.1016/j.beth.2016.04.005
66. Sanchez A, Vazquez C, Gomez D, and Joormann J. Gaze-fixation to happy faces predicts mood repair after a negative mood induction. Emotion. (2014) 14:85–94. doi: 10.1037/a0034500
67. Speirs C, Belchev Z, Fernandez A, Korol S, and Sears C. Are there age differences in attention to emotional images following a sad mood induction? Evidence from a free-viewing eye-tracking paradigm. Neuropsychol Dev Cognit B Aging Neuropsychol Cogn. (2018) 25:928–57. doi: 10.1080/13825585.2017.1391168
68. Raila H, Scholl BJ, and Gruber J. Seeing the world through rose-colored glasses: People who are happy and satisfied with life preferentially attend to positive stimuli. Emotion. (2015) 15:449–62. doi: 10.1037/emo0000049
69. Surber C, Hoepfel D, Günther V, Kersting A, Rufer M, Suslow T, et al. Deployment of attention to facial expressions varies as a function of emotional quality-but not in alexithymic individuals. Front Psychiatry. (2024) 15:1338194. doi: 10.3389/fpsyt.2024.1338194
70. Hoepfel D, Bila A, Günther V, Kersting A, and Suslow T. Attention to facial emotions in adult women varies by type and severity of childhood maltreatment experience and emotion regulation strategy. Sci Rep. (2025) 15:16266. doi: 10.1038/s41598-025-99562-z
71. Isaacowitz DM, Wadlinger HA, Goren D, and Wilson HR. Selective preference in visual fixation away from negative images in old age? An eye-tracking study. Psychol Aging. (2006) 21:40–8. doi: 10.1037/0882-7974.21.1.40
72. Faul L, Bellaiche L, Madden DJ, Smoski MJ, and LaBar KS. Depression and emotion regulation strategy use moderate age-related attentional positivity bias. Front Psychol. (2024) 15:1427480. doi: 10.3389/fpsyg.2024.1427480
73. Fernandes C, Silva S, Pires J, Reis A, Ros AJ, Janeiro L, et al. Eye-tracking evidence of a maintenance bias in social anxiety. Behav Cognit Psychother. (2018) 46:66–83. doi: 10.1017/S1352465817000418
74. Gamble AL and Rapee RM. The time-course of attention to emotional faces in social phobia. J Behav Ther Exp Psychiatry. (2010) 41:39–44. doi: 10.1016/j.jbtep.2009.08.008
75. Soltani S, Newman K, Quigley L, Fernandez A, Dobson K, and Sears C. Temporal changes in attention to sad and happy faces distinguish currently and remitted depressed individuals from never depressed individuals. Psychiatry Res. (2015) 230:454–63. doi: 10.1016/j.psychres.2015.09.036
76. Hatfield E, Cacioppo JT, and Rapson LR. Emotional contagion. New York: Cambridge University Press (1994).
77. Hatfield E, Bensman L, Thornton PD, and Rapson RL. New perspectives on emotional contagion: A review of classic and recent research on facial mimicry and contagion. Interpersona. (2014) 8:159–79. doi: 10.5964/ijpr.v8i2.162
78. Dimberg U, Thunberg M, and Elmehed K. Unconscious facial reactions to emotional facial expressions. Psychol Sci. (2000) 11:86–9. doi: 10.1111/1467-9280.00221
79. Olszanowski M, Wróbel M, and Hess U. Mimicking and sharing emotions: A re-examination of the link between facial mimicry and emotional contagion. Cognit Emot. (2020) 34:367–76. doi: 10.1080/02699931.2019.1611543
80. Guy N, Sklar AY, Amiaz R, Golan Y, Livny A, and Pertzov Y. Individuals vary in their overt attention preference for positive images consistently across time and stimulus types. Sci Rep. (2024) 14:8712. doi: 10.1038/s41598-024-58987-8
81. Rohr M, Degner J, and Wentura D. The ‘emotion misattribution’ procedure: Processing beyond good and bad under masked and unmasked presentation conditions. Cognit Emot. (2015) 29:196–219. doi: 10.1080/02699931.2014.898613
82. Wentura D, Rohr M, and Degner J. Masked emotional priming: A double dissociation between direct and indirect effects reveals non-conscious processing of emotional information beyond valence. Conscious Cogn. (2017) 49:203–14. doi: 10.1016/j.concog.2017.01.016
83. Sears CR, Thomas CL, LeHuquet JM, and Johnson JCS. Attentional biases in dysphoria: An eye-tracking study of the allocation and disengagement of attention. Cognit Emot. (2010) 24:1349–68. doi: 10.1080/02699930903399319
84. Sun R, Fietz J, Erhart M, Poehlchen D, Henco L, Brückl TM, et al. Free-viewing gaze patterns reveal a mood-congruency bias in MDD during an affective fMRI/eye-tracking task. Eur Arch Psychiatry Clin Neurosci. (2024) 274:559–71. doi: 10.1007/s00406-023-01608-8
85. Aldao A, Nolen-Hoeksema S, and Schweizer S. Emotion-regulation strategies across psychopathology: A meta-analytic review. Clin Psychol Rev. (2010) 30:217–37. doi: 10.1016/j.cpr.2009.11.004
86. Suslow T, Hoepfel D, Günther V, Kersting A, and Bodenschatz CM. Positive attentional bias mediates the relationship between trait emotional intelligence and trait affect. Sci Rep. (2022) 12:20733. doi: 10.1038/s41598-022-25317-9
87. Anwyl-Irvine AL, Armstrong T, and Dalmaijer ES. MouseView.js: Reliable and valid attention tracking in web-based experiments using a cursor-directed aperture. Behav Res Methods. (2022) 54:1663–87. doi: 10.3758/s13428-021-01703-5
88. Woronko SE, Jessup SC, Armstrong T, Anwyl-Irvine AL, Dalmaijer ES, and Olatunji BO. A novel probe of attentional bias for threat in specific phobia: Application of the “MouseView.js” approach. J Anxiety Disord. (2023) 96:102700. doi: 10.1016/j.janxdis.2023.102700
89. Milani S, Armstrong T, Dalmaijer E, Anwyl-Irvine A, and Dawson SJ. Examining attentional biases elicited by sexual stimuli using MouseView.js: An online paradigm to mimic eye movements. J Sex Res. (2025) 62:150–63. doi: 10.1080/00224499.2023.2209792
90. Burns H, Hurst A, Garay P, Murray NE, Stewart SH, Mejia J, et al. Attentional biases for dynamic stimuli in emerging adults with anxiety: A preliminary eye-tracking study. J Psychiatr Res. (2025) 184:262–71. doi: 10.1016/j.jpsychires.2025.02.046
Keywords: attention, emotion regulation, eye-tracking, free viewing, positive attentional bias, positive mood, time-course of bias
Citation: Suslow T, Hoepfel D, Wenk T, Kersting A and Günther V (2025) A look at the free-viewing paradigm in eye-tracking research to assess positive attentional bias. Front. Psychiatry 16:1659072. doi: 10.3389/fpsyt.2025.1659072
Received: 03 July 2025; Accepted: 26 November 2025; Revised: 26 November 2025;
Published: 10 December 2025.
Edited by:
Roberto Viviani, University of Innsbruck, AustriaReviewed by:
Mary E. McNamara, McLean Hospital, United StatesCopyright © 2025 Suslow, Hoepfel, Wenk, Kersting and Günther. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Thomas Suslow, dGhvbWFzLnN1c2xvd0BtZWRpemluLnVuaS1sZWlwemlnLmRl