Eye movement pattern of attention bias to emotional stimuli in women with high premenstrual symptoms

Highlights

• HPMS group showed an orientation bias to negative stimuli.

• HPMS group showed difficulty in disengaging attention from negative stimuli.

• HPMS group showed accelerated attentional disengagement from positive stimuli.

• HPMS group had higher trait rumination than LPMS group.

Abstract

Background and objectives

Premenstrual Syndrome (PMS) refers to a group of symptoms linked to the menstrual cycle. Women with PMS have cognitive mode of rumination, which leads to their attention bias to emotional stimuli. This study investigated the biases for emotional information in women with high premenstrual symptoms (HPMS) compared with women with low premenstrual symptoms (LPMS).

Methods

A total of 38 women with HPMS and 44 women with LPMS completed self-report questionnaires and a free viewing task with eye-tracking technology.

Results

The questionnaire results indicate that women in the HPMS group had higher levels of rumination than those in the LPMS group. The eye-tracking results show that women in the HPMS group had an orientation bias towards negative emotional stimuli in the early cognitive process. In the late cognitive process, women in the HPMS group had accelerated attentional disengagement to positive emotional stimuli and difficult attentional disengagement to negative emotional stimuli. Further correlation analysis revealed positive relationships between the scores of initial fixation latency bias of positive pictures in premenstrual phase and the scores of symptom rumination in both groups and between the scores of initial fixation latency bias of positive pictures in premenstrual phase and the scores of brooding in HPMS group.

Limitations

This study used a retrospective questionnaire to assess the symptoms of PMS.

Conclusions

Women with HPMS had impaired attentional engagement and disengagement to emotional stimuli compared with women with LPMS, and it may be related to their cognitive mode of rumination.

Introduction

Premenstrual syndrome (PMS) refers to a series of physiological, emotional, and behavioral symptoms among the women of reproductive age. These symptoms appear periodically in the luteal phase, with remission generally occurring within 3 days after the onset of menses, and interfere with daily functioning (Yonkers, O'Brien, & Eriksson, 2008). PMS is characterized by a combination of symptoms such as irritability, depression, anxiety, and breast tenderness (Qiao et al., 2012).

Studies have found that women with PMS have a cognitive mode of rumination (Craner, 2014; Craner, Sigmon, Martinson, & McGillicuddy, 2014; Craner, Sigmon, & Young, 2015). Rumination is a negative cognitive style, which refers to recurrent negative thinking about the causes and various possible negative consequences, without proactive problem-solving (Nolen-Hoeksema, 1991). Hunter (2003) proposed a cognitive model for women with PMS by arguing that negative cognitive appraisal of premenstrual changes may help explain the pathogenesis of PMS. Studies have shown that women with PMS reported a higher level of trait rumination than women without PMS (Craner, Sigmon, & Young, 2015), and that the level of trait rumination predicted the severity of PMS (Craner, 2014). Participants in the PMS group also showed more ruminant responses to negative emotions than those in the control group in a laboratory setting (Craner, Sigmon, & Young, 2015). The cognitive model of rumination may play an important role in elaborating the cognitive–emotional process in women with PMS (Takano & Tanno, 2008).

Recent studies found that attention biases (ABs) are associated with rumination (e.g., Armstrong &Olatunji, 2012; Kaiser et al., 2018). ABs are defined as a tendency to attend to a certain type of stimulus over others during emotional information processes (Beckwe & Deroost, 2016). ABs are driven by two components: attentional engagement and attentional disengagement (Posner, Inhoff, Friedrich, & Cohen, 1987). Attentional engagement refers to the selection and initial orientation of attention toward stimuli in the early cognitive process, whereas attentional disengagement refers to the withdrawal of attention from stimuli in the late cognitive process (Hollingworth, 2008). ABs may be caused by faster or more frequent initial orientation of attention to related stimuli, or impaired attentional disengagement. The impaired attention disengagement leads to difficulty in diverting the attention away from the related negative stimuli and in the maintenance of related positive (Clarke, MacLeod, & Guastella, 2013). In addition to the static level of attentional engagement and disengagement involved in the ABs, the dynamic processes underlying attentional engagement and disengagement are also involved in the ABs to information, manifested as returning to stimuli repeatedly (Roy-Charland, Plamondon, Homeniuk, Flesch, & Stewart, 2017).

Rumination is characterized by biased attentional processing of negative information (Owens & Gibb, 2016). For example, individuals with higher levels of trait rumination showed more ABs toward negative information (Koster, Derakshan, & Raedt, 2011). Studies have also shown that rumination and ABs are interrelated and affected by mood status such as anxiety and depression. High-anxiety individuals had stronger ABs under the condition of inducing rumination than that of inducing distraction, whereas low-anxiety individuals showed no ABs (Essmann & de Jong-Meyer, 2008). Depressed individuals with a high level of trait rumination were associated with difficulty in attentional disengagement from depression-related information (Grafton, Southworth, Watkins, & MacLeod, 2016). The difficulty of disengagement from depression-related information was found to be associated with trait rumination, but not state rumination (Donaldson, Lam, & Mathews, 2007). Armstrong and Olatunji (2012) reported that anxious individuals showed increased initial attention to threat stimuli, but no difficulty to disengage from threat stimuli. Conversely, depressed individuals did not show any increase in the initial attention to threat stimuli. Although these studies only focused on the static level of ABs, they all revealed that the affecting disorders were associated with different characteristics of altered ABs, which were closely related to rumination.

One most recent study further investigated ABs with regard to the emotional information of women with PMS (Eggert, Kleinstäuber, Hiller, & Witthöft, 2017). Researchers used the emotional Stroop paradigm with emotional words, pictures, and faces and found that participants in the PMS group had ABs toward negative emotional information. Compared to those in the control group, participants in the PMS group showed greater emotional Stroop effect to picture and facial stimuli. However, no significant difference was observed between the PMS and control groups on Stroop effects to word stimuli. With respect to the facial stimuli, researchers also found a type of paradox effect (Stroop facilitation) in the PMS group; i.e., emotional stimuli accelerated the speed of cognitive processing. An earlier study indicated that emotional Stroop tasks could not distinguish between the attentional engagement or disengagement of ABs (Jansen, Nederkoorn, & Mulkens, 2005). It is unclear whether the contradiction of the Stroop effect to different stimuli can be attributed to the acceleration of attentional engagement or the impaired attentional disengagement. Because the traditional AB paradigms heavily rely on the speculative cognitive process by measuring the reaction time; the components and the process of ABs cannot be directly observed. Additionally, it also ignores the dynamics of attentional processing involved in the ABs (Roy-Charland et al., 2017). Because eye-tracking technology has been used in the research of attentional engagement and allows for direct and continuous measurements of overt visual attention, researchers recently proposed that eye-tracking technology could provide an important supplement to reaction time measures (Armstrong, Sarawgi, & Olatunji, 2012).

Based on previous research, this study used the paired emotional picture-viewing paradigm with eye-tracking technology to examine the components, process, and dynamics of emotional information's ABs by comparing women with high premenstrual symptoms (HPMS) and women with low premenstrual symptoms (LPMS). As mentioned above, the symptoms of PMS include both depression and anxiety. Therefore, we hypothesized that the eye movement patterns in women with HPMS would be different in patients with anxiety and depression. We further hypothesized that, compared with women with LPMS, women with HPMS would have accelerated attentional engagement toward negative emotional stimuli and reduced attentional engagement in the orientation to positive stimuli in the early cognitive process. However, they would have difficulties in attentional disengagement from negative emotional stimuli and in the maintenance of positive stimuli in the late cognitive process. Moreover, women with HPMS would return their attention back to negative stimuli more frequently than those with LPMS. Besides, ABs were associated with the cognitive mode of rumination in women with HPMS. The differences in attentional patterns would occur only during the premenstrual period, not in the postmenstrual period.