probiotics
Bachelor thesis, Universiteit van Amsterdam
Merel Schouten
11241519
Duba Psychobiology/Psychology
Supervisor: Laura Steenbergen
22-05-2020
Abstract
Probiotics interventions are proposed to prevent and treat psychological disorders, as
depression, by modulating the microbiota-gut-brain axis. The neural connection between the gut microbiome and the brain presupposes that probiotics intervention might alter emotion processing. Disrupted emotion processing has been presented as an accountable domain fundamental to depression and many other psychological disorders resulting in impaired recognition of emotional expressions. The present study aimed to test the effect of a 4-week probiotics intervention on ability to recognize facial emotion using a placebo-controlled, randomized, pre- and post-intervention assessment design. It was hypothesized that this effect would be moderated by the ability to regulate emotion. Based on the role of interoception in emotion regulation, participants were separated in two groups (high vs. low interoceptive awareness) using pre-intervention MAIA-scores to test this hypothesis. The ability to recognize facial emotion was assessed by determining the 50% emotion identification threshold (i.e. facial emotion intensity at which 50% of the presented emotion expressions were labeled correctly). Emotion identification threshold significantly decreased for
participants with a high ability to regulate emotion who received the probiotics intervention. This indicates increased ability to identify facial emotion. Remarkably, emotion identification threshold also significantly decreased for participants with a low ability to regulate emotion who received the placebo intervention, but not for those who received probiotics. These results provide evidence that the ability to recognize facial emotion may be altered by
manipulation of the gut microbiome with probiotics. Dynamic changes in emotion processing during the 4-week intervention period might explain the aberrant results found for participants with low ability to regulate emotion. Future research should specify an “antidepressive effect” regarding emotion regulation to clarify the implications of probiotics as a potential
intervention for depression.
Keywords: Probiotics, emotion perception, identification threshold, interoception, emotion
Introduction
Our gut hosts a great number of commensal microorganisms and their genomes (i.e. the gut microbiome). Bacteria dominate the gut microbiome with thousands of species and strains (Lozupone et al., 2012). In a healthy functioning gut microbiome, there is a balance between beneficial and pathogenic bacteria (Mallick-Searle, 2019). The microbiome is accessible for environmental factors since the gastrointestinal tract is essentially an open system. Various bacteria, nutrients, yeasts, antigens and drugs that pass through the intestine after ingestion can disrupt the balance between good and pathogenic bacteria in the gut microbiome. Diet, consumption of antibiotics, stress mediated physiological effects (e.g. nausea), age and/or change in habitat can (temporarily) alter the composition of the gut microbiome. Therefore, microbiome composition is the result of the condition of the gut ecosystem, recent exposure, diet and fitness of the host (Derrien & Van Hylckama Vlieg, 2015). In chronic imbalanced microbiome conditions, the balance between beneficial and pathogenic bacteria is disturbed. Such an imbalanced microbiome is associated with serious issues ranging from obesity and inflammatory diseases (Claesson et al., 2012) to behavioural (Sivamaruthi et al., 2019) and neuropsychiatric problems including depression (Cenit, Sanz, & Codoñer-Franch, 2017).
The apparent role of microbiome conditions in (mental) health elicited interest to investigate interaction pathways fundamental to the impact of gut bacteria on health status. It was found that the gut microbiome is bidirectionally connected to the brain by multiple indirect communication pathways, generally described as the gut-brain axis (Bonaz, Bazin & Pellissier, 2018). The system connecting the microbiome, gut, and brain includes the enteric nervous system [ENS], the sympathetic and parasympathetic divisions of the autonomic nervous system [ANS], as well as neuroimmune and neuroendocrine components of the central nervous system [CNS] (Ross et al., 2019; Sarkar et al. 2018; Cryan & Dinan, 2012). Gut microbiota act on all of those systems by contributing to production of short-chain fatty acids [SCFA’s] and neuro metabolites (Yano et al., 2015), modulation of the HPA-axis (Sudo et al., 2004) and stimulation of enteroendocrine cells [EEC] to release hormones (Raybould, 2010). Those bacterially generated signals modulate psychophysiological functions
fundamental to psychological processes such as emotion, cognition and social behaviour (Cryan & Dinan, 2012; Rieder et al., 2017; Sarkar et al., 2018). The vagal nerve [VN] is a principle component of the parasympathetic nervous system [PNS] connecting the ENS to the
CNS with 80% afferent and 20% efferent fibers. The VN has a central role in transferring gut microbiota signals to the brain and vice versa (Bonaz et al., 2018). Viscero-sensory inputs from the afferent fibers of the VN contain information about the internal state of the body. Those interoceptive signals mediate to widespread areas of the brain associated with affective disorders (Zagon et al, 2001).
Involvement of psychologists in microbiological research on the gut-brain axis raised interest in the role of the gut microbiome in psychological disorders. Also, design of strategies to modulate the gut microbiota and its functions to prevent and treat psychological disorders became a point of interest (Mayer et al. 2014; Bastiaanssen et al. 2020). Novel insights promote the hypothesis that modulation of the gut microbiome with probiotics supplements might serve as interventions in psychopathology (Sarkar et al., 2018; Wieërs et al., 2020). Probiotics do not reshape the commensal microorganisms nor do they permanently modify the microbiota composition or diversity. Yet, they support challenged microbiota, influence epithelial and immune cells and share genes and metabolites in the gut (Wieërs et al., 2020). The psychological implications of the physiological effects of probiotic interventions remain largely untested. According to the definition of the World Health Organization [WHO], probiotics are “live microorganisms which, when administered in adequate amounts, confer a health benefit on the host.” (Food and Agriculture Organization of the United Nations [FAO] & World Health Organization [WHO], 2001). Consumption of probiotics as a food ingredient or a dietary supplement is very common. In fact, probiotic supplements are one of the most commonly used nutritional supplements worldwide nowadays (Abid & Koh, 2019). If supporting the gut microbiome with probiotics can contribute to prevention and treatment of psychological disorders, implications will not be limited to management of (neuro)psychiatric diseases and behavioural problems. Elaboration of knowledge on the impact of probiotic intervention will also contribute to the understanding of general health benefits experienced by healthy people consuming probiotic supplements on a day to day basis.
Probiotic interventions have been found to reduce corticosterone levels induced by stress and reduce behavioural depression- and anxiety- symptoms in healthy animals (Bravo et al., 2011). In humans the consumption of probiotics has also been associated with reduction in self-reported depression and anxiety symptoms (Messaoudi et al., 2011). Additionally, it has been related to decrease in negative mood in volunteers reporting negative mood at baseline (Benton, Williams & Brown, 2007). Further research focussed on the impact of probiotics intervention on dysfunctional cognitive processing found predictive for stress- and
emotion-related disorders like anxiety and depression (Beck et al, 1967; Ingram, Miranda & Segal, 2006). Steenbergen et al. (2015) showed that a 4-week multi-strain probiotics
intervention reduced rumination and aggressive reactivity to negative thoughts in healthy participants not diagnosed with mental disorder. Bagga et al. (2018) also found reduced cognitive reactivity to sad mood; reduction in self-reported measures of hopelessness and risk aversion. The aforementioned studies have taken the first steps unravelling bacterial
contributions to psychological processes studying the behavioural and cognitive impact of probiotic interventions. Physiological decrease in cortisol response, reduction in negative mood and self-reported affective changes after probiotics intervention may have one
underlying neurocognitive mechanism according to Sarkar et al. (2018): a general reduction in physiological reactivity to emotional stimuli, positive and negative.
The hypothesised reduction in emotional reactivity sheds light on the affective domain of psychology, a depreciated domain in the study of bacterial contribution to psychological processes thus far. Disrupted emotion processing has been proposed as a burden for
comorbidity of mental health disorders. Emotion processing is a complex of affective, behavioural, and cognitive mechanisms underlying our emotions. Deficits in emotion
processing, as reported in multiple psychological disorders, result in impaired recognition of emotional expressions at the perceptual level (Kret & Ploeger, 2015). Basic facial expressions (i.e. anger, disgust, fear, happy, sad, and surprise) are uniformly expressed by all people of all cultures and may even be biologically hardwired according to Darwin (Frank, 2001). The recognition of those basic emotions is related to the ability to interpret the feelings and emotions of others which is important for social interaction (Bomfim, Ribeiro & Chagas, 2019). The therapeutic actions of antidepressant medications have been found to modulate emotional processing in treatment of depressed patients, even before changes in mood and symptoms (Harmer et al., 2009). Experimental psychology studies focussing on the effects of probiotics interventions on emotion processing are missing despite interest in probiotics supplementation as an intervention in psychopathology. The current study aims to partly fill this knowledge gap by looking at the direct effect of a 4-week multispecies probiotic
intervention on emotion perception.
The neural system for recognition of emotion consists of forebrain structures (i.e. the amygdala and basal ganglia) and cortical structures like the occipitotemporal neocortex, orbitofrontal cortex and right parietal cortices (Adolphs, 2002). When facial expressions are
presented, a subcortical route via the superior colliculus and the pulvinar thalamus activates the amygdala and other limbic system structures (Young et al. 2007). The first responses of the human amygdala to emotional facial expressions are seen after 120 ms, even when the facial expressions are shown subliminal. (Adolphs, 2002). All of those brain regions are innervated by the nucleus tractus solitarius [NTS]. The NTS is a brainstem network that receives visceral input from the afferent fibers of the VN. Sellaro et al. (2018) demonstrated that active stimulation of the afferent vagal nerve fibers with transcutaneous vagus nerve stimulation [tVNS] improves recognition of facial emotion. Activation of the VN by
positively stimulating the gut microbiome with a probiotics-intervention may have the same impact (Perez-Burgos et al., 2013; Gaykema, Goehler & Lyte, 2004).
Brain activity measures revealed reduced activity in brain regions processing emotion after a 4-week intervention with a fermented milk product containing probiotics (Tillisch et al., 2013). This result is in line with the hypothesis that probiotics reduce physiological emotional reactivity as proposed by Sarkar et al. (2018). However, Bagga et al. (2018) showed that a 4-week probiotics intervention enhanced attention to emotional stimuli in an emotional decision-making task and an emotional recognition memory task with negative and neutral images. Lastly, Kelly et al. (2017) studied the effect of probiotics intervention on recognition of facial emotion specifically. No significant changes in accuracy and reaction time during an Emotion Recognition Task (ERT) in an 8-week cross-over design were found. This limited number of studies, with varied results, illustrate the need for an experimental psychology study examining effects of probiotic intervention on the ability to recognize facial emotion. First of all, to learn about the impact of probiotics intervention on emotion
processing and eventually to study the implications of those results for probiotics-intervention as potential contribution to prevention and treatment of psychological disorders.
Remarkably, studies with acknowledged medication for treatment of psychological disorders show divagating results regarding the effects on emotion processing (review: Merens et al., 2007). The Emotional Test Battery [ETB] is commonly used for experimental assessment of emotion processing (Murphy et al., 2008; Thomas et al., 2016). This test battery comprises five tasks; The Facial Expression Recognition Task [FERT], Faces Dot Probe Task [FDOT], Emotional Categorisation Task [ECAT], Emotional Recall Task and Emotional Recognition Memory Task [EMEM]. A number of studies found that anti-depressant intake improved ability to recognize positive facial expressions and/or deteriorated recognition accuracy for negative facial emotions in healthy volunteers (Harmer et al., 2003a; Harmer et
al., 2004; Murphy et al., 2006). Contrary, Horder et al. (2009) found that antidepressants only decreased positive emotional memory without effecting other tests from the ETB. Yet, Harmer et al. (2003b) and Harmer et al. (2008) found that acute administration of
antidepressant to healthy volunteers resulted in a higher number of correctly identified fear (negative) and happy (positive) expressions. Coupland et al. (2003) also found a broader effect of antidepressants on emotion recognition than specifically on fear and happy facial expressions. They found that administration of anti-depressant to healthy participants directly increased the identification threshold (i.e. the lowest intensity at which each emotion stimulus was recognized correctly without subsequent alteration) for recognition of all emotional faces. However, the results were only significant for surprised faces. The increased identification threshold indicated a significant depletion in ability to recognize surprised facial emotion expression. Murphy et al. (2008) used a lower dose of antidepressant on healthy volunteers compared to the studies described above. No effect on mood, alertness, sustained attention, accuracy or reaction time in the FERT, nor an effect on ECAT, emotional recall and EMEM was found. To summarize, effects of anti-depressant administration on the ability to recognize facial emotion are not always observed. In case such an effect is observed, they point towards different directions of change. It remains inconclusive whether lowering or increasing the ability to recognize facial emotion has an anti-depressive impact.
For the current study the question is whether probiotics-intervention lowers or
increases the ability to recognize facial emotion in healthy volunteers. Probiotics are found to manipulate the serotonergic system similar to the antidepressant interventions discussed above (Yano et al., 2015). Yet, even when implying that probiotics would also have the same impact on affective processing, results from these studies do not pointing towards one specific expected direction of change.
If probiotics increase the ability to recognize facial emotion expressions, this will most likely positively contribute to effective social interaction as it improves your ability to
interpret the feelings and emotions of others (Bomfim et al., 2019). However, in order to benefit from this, you have to protect your own mental and physical health by successfully regulating emotional reactivity (Kret & Ploeger, 2015). One should be able to down- or up-regulate emotion according to social conventions or personal attitudes.
Emotion experience and the ability to regulate emotion is necessarily associated with awareness of one’s emotional and bodily state (Füstös et al., 2013; Critchley & Harrison, 2013). The total autonomic sensory nerve input from the body, with which you sense the
physiological condition of the material body, is called interoception (Craig, 2002). The limbic system, specifically the amygdala, is part of the emotion processing brain network and
relevant for interoception (Cameron, 2001). Interoception is therefore associated with activity of the emotion processing network innervated by the VN via the NTS as described earlier, and may hence be affected by the intake of probiotics (Young et al., 2007; Tillisch et al., 2013). Sensitivity to interoceptive information is positively related to the intensity of affective response towards emotion stimuli (Jones, 1994; Montoya and Schandry, 1994; Cameron, 2001; Critchley et al, 2004; Herbert et al., 2007) and to regulation of emotional reactivity (Critchley & Harrison, 2013; Füstös et al., 2013; Price & Hooven, 2018; Zamariola et al., 2019). Interoceptive sensitivity varies widely among individuals (Wiens, Mezzacappa & Katkin, 2000). The necessary role of interoception in emotion experience and regulation implicates that this variation is also displayed in the ability to regulate emotion. Volunteers with differences in the ability to regulate emotion may not undifferentiated profit from lowering or increase in the ability to recognize facial emotion.
Given the aforementioned, it is hypothesised that the ability to regulate emotion is moderating the effect of probiotics on facial emotion recognition (Figure 1).
Specifically, it is expected that participants with a high ability to regulate emotion, show improved ability to recognize emotional faces after a 4-week intervention with multispecies probiotics. Because of their ability to protect their own mental and physical health with successful emotion regulation, they can benefit from the improved ability to recognize facial emotion expressions. This positively contributes to effective social
interaction. Unless the ability to regulate emotion increases, participants with a low ability to regulate emotion are expected to show a decreased ability to recognize emotional faces after a 4-week intervention with multispecies probiotics. This decreased ability to recognize facial emotion will offer “protection” against emotional reactivity, which they are not able to control successfully. Without such “protection” increased ability to recognize may result in mental of physical harm.
To investigate whether the effects of probiotics on emotion perception are indeed moderated by the ability to regulate emotion, the participants in this study will be separated in two groups based on self-reported interoceptive awareness. The Multidimensional
Assessment of Interoceptive Awareness (MAIA) is a self-report measure for interoceptive body awareness. A high total score indicates a higher level of interoceptive awareness
(Mehling et al., 2012). This higher level of interoceptive awareness is again related to a greater ability to regulate emotion (Remmers et al., 2016; Zamariola et al., 2019).
Figure 1. Expected effects of a 4-week multispecies probiotics intervention on ability to recognize facial emotion with ability to regulate emotion, represented by MAIA-score, as a moderating factor.
Materials and method
Participants
Participants were recruited through online advertisements on social media and word-to-mouth. The participants in this study were 89 healthy young adults (29 male) between 16 and 35 years old (mean = 22.26, SD = 3.09). They reported no prescribed medication or drug use (except of hormonal contraceptives), no use of antibiotics or probiotics in the past 3 months, no participation in an intensive dietary program for health reasons or foreseen to need gastrointestinal surgery during the study period and no current pregnancy or breast-feeding or intention to become pregnant during the study (Bagga et al., 2018). Participants who reported no gastrointestinal or cardiovascular disease, no mental or physical disability that would hinder participation in this intervention, no current suffering from psychological/psychiatric
disease, no hypersensitivity or allergy to milk protein, soy protein or gluten and no history of cancer were considered applicable to the study (Benton et al., 2007; Steenbergen et al., 2015). Besides, all participants reported to not consume more than 20 units of alcohol per week and no use of soft- or hard drugs more than once a month (Messaoudi et al, 2011).
Participants were randomly and equally assigned to either a placebo or a probiotics intervention lasting for four weeks. A randomization list, generated by the web site
Randomization.com (Dallal, 2007), was used to assign each participant to one of the
interventions. This was done using the method of randomly permuted blocks (Appendix A). The ascription of participants to one of the two conditions was therefore blinded at three levels; group allocator, participants and outcome assessor.Before the start of the study a written informed consent was obtained from all of the participants and the protocol, registered as CEP19-0311194, was approved by the local ethical committee (Leiden University, Institute for Psychological Research).
Design
To investigate the effect of a multispecies 4-week probiotics intervention on facial emotion perception a placebo-controlled, randomized, pre- and post-intervention assessment design was used. In both conditions (probiotics and placebo), participants received a food supplementation intervention in the form of 28 identical sachets with 2g of freeze-dried powder (one for each day of the 4-week intervention). During the intervention the content of these sachets was exogenously added to the gut microbiome via ingestion. In the probiotics condition the sachets each contained the probiotics mixture Ecologic®Barrier 849 (Winclove probiotics, The Netherlands). In the placebo condition the sachets only contained the carrier of the probiotics product and not the bacterial strains. However, the sachets were
indistinguishable in color, taste and smell.
The pre- and post-intervention assessments were the same for all participants. The complete test battery lasted about two hours in each session (pre- and post-intervention). The pre-intervention assessment started with assessing background information on dietary habits, perceived general fitness, age, use of intoxicants, food supplements and/or antibiotics, (family) medical history, health of cardiovascular system, respiratory system, stomach, intestines and kidneys, functioning of nerve system and sensory system. For female participants, a question on the use of contraceptives and chance of pregnancy was also included in the background questionnaire. Next, BMI variables (length and weight) and the width of the waist and hips were measured. Before the participants started filling out the first
five questionnaires presented in Qualtrics software (Qualtrics, Provo, UT), they were asked to put on a polar heart rate monitor chest belt. The first two questionnaires asked to report on experiences with bowl complains (e.g. bloating, stomach pain etc.) and defecation
frequency/consistency (Bristol Stool Scale). In addition, the participants filled out the Positive and Negative affect scale (PANAS; Watson et al., 1988), the Bermond-Vorst Alexithymia Questionnaire (BVAQ; Vorst & Bermond, 2001) and the Multidimensional Assessment of Interoceptive Awareness (MAIA; Mehling et al., 2012). Following these first five
questionnaires the participant’s resting state heart rate variability (HRV) was monitored for 5 minutes using the polar heart rate monitor chest belt and the Elite HRV Heart Rate Variability app. Next, the participants filled out five more questionnaires; the Perceived Stress Scale (PSS; Cohen, 1994), the Centre for Epidemiological Studies – Depression scale (CES-D, Radloff, 1977), the State-Trait Anxiety Inventory (STAI; Spielberger, 1983), the Emotional Reactivity Scale (ERS; Nock et al., 2008) and the Aggression Questionnaire (AQ; Buss & Perry, 1992). After filling out these questionnaires the participants performed multiple tasks on the computer using the computer mouse and the keyboard to respond. This series of tasks lasted about one hour and twenty minutes and measured emotion-specific and general cognitive emotional biases. The tasks were presented in the E-prime 2.0 software system (Psychology Software Tools, Inc., Pittsburgh, PA)in the following order: The Emotional Categorization Task (ECAT; Harmer et al, 2004;2009), the Faces dot-probe task (adapted from Murphy et al., 2008), the Facial Expression Recognition task (FERT; Harmer et al, 2004;2009, Murphy et al., 2008), the Emotional Recall task (EREC; Harmer et al, 2004;2009, Murphy et al., 2008) and the Emotional recognition Memory task (EMEM; Murphy et al., 2008). Followed by the Scrambled Sentences task (SST; Everaert, Duyck & Koster, 2014) and the Attentional Network Test (ANT; Fan et al., 2002). After completion of the tasks, the participants were asked to fill out two more questionnaires; the Penn State Worry
Questionnaire (PSWQ; Meyer et al., 1990) and the Leiden Index of Depression Sensitivity-revised (LEIDS-r; Van der Does & Williams, 2003)
After the pre-intervention session participants were provided with a carboard box containing 28 laminated aluminum bags (PET/ALU/PE) containing placebo or probiotics that had to be taken in over the course of the 4-week intervention. The participants were instructed to take in the content of one of the sachets every day, using their own supplies, by dissolving it in a glass of luke warm water or milk and drinking this on an empty stomach. During the 4-week intervention participants were asked to fill out a short daily questionnaire, sent by the experimenter, in which they had to answer two (i.e. positive- and negative) feelings questions
and the Bristol stool chart (Lewis & Heaton, 1997). This daily message from the experimenter also facilitated compliance to take the probiotics/placebo on a daily basis.
The post-intervention assessment started with filling out the Empathy Quotient (EQ; Baron-Cohen & Wheelwright, 2004), instead of the background questionnaire from the pre-intervention assessment. Beside this, the post-pre-intervention session was exactly the same as the pre-intervention session. Due to measures of the Dutch government after outbreak of the COVID-19 pandemic in the Netherlands, Leiden University had to close al laboratory
facilities after March 14 2020. All post-intervention sessions after this date had to be adjusted to an online test session. Participants were asked to fill out all the questionnaires from the original post-intervention session from home using their own computer devises. The tasks (ECAT, FERT, EREC/EMEM, SST and ANT) were no longer part of this online post-intervention assessment. The participants received instructions on the course of this online version of the post-intervention assessment by e-mail. They were asked to mimic the original lab situation as much as possible by sitting in a quiet spot in their home during the intervention session. All questionnaires had to be finished on the original date of the post-intervention assessment in the lab, exactly four weeks after the pre-post-intervention assessment. In case of trouble with accessing or answering the questionnaires, participants could contact the test leader by phone. This change in the protocol during the study, due to circumstances beyond the principle investigators control, was approved by the local ethical committee (Leiden University, Institute for Psychological Research). After the post-intervention session, participants were debriefed and rewarded for their participation with 30 euros (7.50 per hour).
Probiotics and placebo
The probiotics mixture Ecologic®Barrier 849 ( 2.5 x 109) contains the following
bacterial stains; Bifidobacterium bifidum W23, Bifidobacterium lactis W52, Lactobacillus acidophilus W37, Lactobacillus brevis W63, Lactobacillus casei W56, Lactobacillus
salivarius W24, Lactococcus lactis (W19) and and Lactococcus Lactis (W58). Those genera belong to the lactic acid bacteria (LAB) also used for fermentation of foods and beverages and considered save to consume (Yadav, Mandeep & Shukla, 2019). See appendix B for more in depth information on the Ecologic®Barrier 849 probiotics mixture. The carrier of the
probiotics in the sachets consist of; Maize starch, Maltodextrins, Vegetable protein, Potassium chloride and Manganese sulphate. In the placebo condition the sachets only contained this carrier of the probiotics product and not the bacterial strains.
Facial emotion recognition task (FERT)
The Facial expression recognition task (FERT; Harmer et al, 2004;2009, Murphy et al., 2008) featured five basic emotions (happiness, sadness, fear, anger and disgust) taken from ten individual faces from the Pictures of Facial Affect series (Ekman & Friesen, 1976). Earlier research also displayed “surprise”, yet “fear” was in most cases confused with “surprise” (Kemmis et al., 2007). Therefore, we excluded “surprise” in this study, similar to de Groot and Steenbergen (2019) to prevent confusion. The emotional faces were morphed between each prototype and neutral by taking a variable percentage of the shape and texture differences between the two standard images 0% (neutral) and 100% (full emotion) in steps of 10% (Young et al., 1997). All five emotions were therefore presented in 10 different
intensities from 10% till 100%. For every emotion four examples (2 actors x 2 presentations) at each of 10 intensities were presented (5 emotions x 10 intensities). Per emotion 1 neutral face at 100% intensity was shown resulting in a total of 204 trails. The faces were shown on a computer screen in random order for 500 ms each and replaced by a list which connected the five emotions and neutral expression to the numbers 1 till 6. The participants had to indicate which emotion they thought the face depicted by pressing one of the numbers 1 till 6 on the keyboard. Since reaction time (in msec) and accuracy was measured in this task, participants were asked to respond as quickly and as accurately as possible. Before the start of the actual task 12 practice trials were presented. Instead of facial emotions, Dutch words indicating one of the five emotions or the word “neutraal” (Dutch for “neutral”) were displayed on the screen. To familiarize with the response keys participants had to match those Dutch words to the English emotion-labels connected to the numbers 1-6 on the keyboard. Between all trials a central fixation cross was displayed for one second. The total test took on average 10 minutes to complete.
The percentage of correct responses was determined for all of the 10 different emotion intensities, taking together the 5 basic emotion presented. As the neutral facial expression indicates no emotion and is only displayed at 100% intensity, the accuracy on those 4 trials was not included. Based on these accuracy measures the emotion intensity at which 50% of the emotional faces were correctly connected to the emotion-labels (i.e. 50% emotion identification threshold) was determined. Determining such an identification threshold using systematically varied emotion intensities is a more graduate way to report findings on manipulation of emotion recognition compared to raw accuracy or reaction time scores used in previous studies (Kemmis et al., 2017, de Groot & Steenbergen, 2019). A lowering of the identification threshold indicates an improvement in perception of emotional faces. To assess
the 50% emotion identification threshold, a linear regression (% correct = a*intensity + b) was computed for each participant and each session (pre- and post- intervention) separately. All coefficients (a) and constants (b) were saved to compute the 50% emotion identification threshold for all participants in each of the sessions (pre- and post-intervention) using the following function: (50 – b)/a.
Multidimensional Assessment of Interoceptive Awareness (MAIA)
The Multidimensional Assessment of Interoceptive Awareness (MAIA; Mehling et al., 2012) is a self-report measure for interoceptive body awareness. The 32-item questionnaire consists of eight separately scored subscales; noticing, not-distracting, not-worrying, attention regulation, emotional awareness, self-regulation, body-listening and trusting. All of the subscales comprise three to seven items. The items are statements like: “I listen to information from my body about my emotional state” and “I notice that my body feels different after a peaceful experience”. All items are self-scored on a scale from 0 (never) to 5 (always). The items 5,6,7,8 and 9 had to be reverse coded to make sure a higher score on the subscale indicates a higher level of awareness for all subscales. The eight subscales reflect five overall dimensions; awareness of body sensations, emotional reaction and attentional responses to sensations, capacity to regulate attention, awareness of mind-body integration and trusting body sensations. All of the dimensions are represented by up to three of the subscales. This study focused on the total MAIA-scores obtained by summing the (recoded) self-reported scores on all 32 items of the questionnaire for all participants. A higher total score indicates a higher level of positive interoceptive awareness.
Statistical analysis
In line with the interest of this study, participants from both intervention conditions were categorized into a group with a high ability to regulate emotion and a group with a low ability to regulate emotion. To address emotion regulation at baseline, focus was on pre-intervention MAIA-scores. The MAIA does not come with pre-defined scores categorizing the continuous MAIA-score into separate ability levels. Therefore, a median split approach was used to separate all participants into two artificial categories (low vs. high
pre-intervention MAIA-score).
Total MAIA-scores were submitted to a mixed analysis of variance [ANOVA] with time (pre- and post-intervention) as within-subjects factor and both intervention-group (probiotics/placebo) and ability to regulate emotion at baseline (low MAIA/high MAIA) as
between-subject factors to see whether MAIA-scores changed over the course of the 4-week intervention. The 50% emotion identification threshold values were computed using a linear regression function and submitted to a mixed ANOVA with the same within- and between-subject factors.
The assumption of normality was checked using the Shapiro-Wilk test and analysis of the QQ-plots before conduction of the mixed ANOVA analyses. Raw data was transformed if the assumption of normality was violated. Mauchly’s Test of Sphericity was of no
information since pre- and post-intervention measures result in only one difference score. Bonferroni corrected pairwise comparison was used to clarify differences in case of significant main- or interaction effects. For all analyses alpha levels were set at p = .05.
Results
Participants
A total of 91 participants gave consent to the course of the study as described in the information letter they received. Upon filling out the background check 2 participants
appeared not to meet the criteria for the study despite screening. Both were sent home before starting the pre-intervention test session. A total of 44 participants (15 male) with a mean age of 22.25 (SD = 3.07) were assigned to the probiotics condition, and 45 participants (14 male) with a mean age of 22.27 (SD = 3.15) were assigned to the placebo condition (Table 1). Analysis of the demographic characteristics of all eighty-nine participants revealed no significant differences for age (t(87) = 0.03, p =0.98) or gender (χ2 (1; N = 89) = 0.09, p =
0.76) between the two intervention groups (probiotics vs. placebo). Table 1.
Demographic characteristics participants for the probiotics and placebo intervention group. Mean age is displayed and standard deviations are shown within parentheses.
Probiotics Placebo Total
Number of participants (M:F)
44 (15:29) 45 (14:31) 89 (29:60)
Age in years 22.25 (3.07) 22.27 (3.15) 22.26 (3.09)
Two participants dropped out after the first session in the lab. Due to technical error FERT data of the pre- and or post-intervention of four participants was not recorded. After outbreak of the COVID-19 pandemic Leiden University had to close al laboratory facilities following Dutch government measures. Therefore, all post-intervention sessions scheduled
after March 14 2020 were adjusted to online sessions. Those online post-intervention sessions did not include the FERT (see Procedure for details). These adjustments affected nineteen participants, causing additional missing post-intervention FERT data. Listed deletion of missing data resulted in excluding 25 participants from all analyses of this study (Figure 2). Hence, analyses of the total scores on the Multivariate Assessment of Interoceptive
Awareness (MAIA) and the data from the Facial Emotion Recognition Task (FERT) were performed on 64 participants (22 male) with a mean age of 22.47 (SD = 3.49).
Figure 2. Flowchart of data collection. Reasons for exclusion are given for all participants excluded from analyses.
A median split approach was conducted on pre-intervention MAIA-scores to separate the 64 participants included for further analysis into two categories (high ability to regulate emotion vs. low ability to regulate emotion). Pre-intervention MAIA-scores ranged from 49 till 143 [mean = 92.89, SD = 18.09] with median 93.5. All participants with a pre-intervention MAIA-score ≥ 93.5 were assigned to the high MAIA group (interpreted as high ability to regulate emotion) and all participants with a pre-intervention MAIA-score < 93.5 were assigned to the low MAIA group (interpreted as low ability to regulate emotion). The participants in the high MAIA group had a significant higher total MAIA score at baseline compared to the low MAIA group [t(62) = 10.37, p < 0.001]. The two intervention conditions (probiotics vs. placebo) were equally represented in both MAIA-groups (high MAIA vs. low MAIA) [χ2(1; N =64) = 1.564, p = 0.211)].
Based on this MAIA-score grouping combined with the two different intervention conditions (probiotics vs. placebo) participants were split into four groups for further data analyses: low baseline ability to regulate emotion + probiotics intervention (N = 13), low baseline ability to regulate emotion + placebo intervention (N = 19), high baseline ability to regulate emotion + probiotics intervention (N = 18) and high baseline ability to regulate emotion + placebo intervention (N = 14). Analysis of the demographic characteristics of all sixty-four participants in those four groups revealed no significant differences for age [F(3,63) = .843, p = .476] or gender [χ2 (3; N = 64) = 1.75, p = 0.432].
MAIA
The variable “MAIA-score” (i.e. the total score on the MAIA questionnaire) was normally distributed except for the pre-intervention MAIA-scores for participants with a low baseline ability to regulate emotion assigned to the placebo intervention [W(19) = .883, p = . 024]. However, the QQ-plot of the MAIA-scores is this category showed no disturbing deviation from a normal distribution. Table 2 displays the mean MAIA-scores values pre- and post-intervention for both intervention conditions and MAIA-groups.
Table 2.
Mean MAIA-score and standard deviation (shown in parentheses) for probiotics and placebo intervention in the low intervention MAIA (low ability to regulate emotion) and high pre-intervention MAIA (high ability to regulate emotion) groups. Asterisks indicate significant between group differences.
Pre-intervention MAIA Post-intervention MAIA Low MAIA pre-intervention Probiotics (N = 13) 79.23 (9.49) 82.77 (12.82) Placebo (N = 19) 78.16 (13.06) 84.37 (18.15) Total * (N = 32) 78.59 (11.58) 83.72(15.99) High MAIA pre-intervention Probiotics (N = 18) 104.72 (77.71) 102.00 (13.56) Placebo (N = 14) 110.36 (12.78) 105.07 (18.07) Total * (N = 32) 107.19 (10.45) 103.34 (15.50) *p < 0.05
The mixed ANOVA, with time (pre- vs. post-intervention) as within-subjects factor and both intervention group (probiotics vs. placebo) and MAIA-score group (low MAIA vs. high MAIA) as between-subject factors, revealed no significant main effect of time on the MAIA-score F(1,60) = .081, p = .777, ηp2 = .001. Yet, a significant time by MAIA-group interaction F(1,60) = 8.452, p = 0.005, ηp2 = .123 was revealed. Bonferroni corrected pairwise comparisons showed that the MAIA-score significantly increased from
pre-intervention [mean = 78.59, sd = 11.58] to post-pre-intervention [mean = 83.72, sd = 18.15, p = . 028] in the group with a low MAIA score at baseline. While the MAIA-score
non-significantly decreased from pre-intervention [mean = 107.19, sd = 10.45] to post-intervention [mean = 103.34, sd = 15.50, p = .067] in the group with a high MAIA score at baseline (Figure 3). No further significant sources of variance were observed.
Figu re 3. Plot significant “time” by “MAIA-group” interaction effect. Mean MAIA-score for both MAIA-score groups pre- and post-intervention. Error bars display 95% confidence-interval.
FERT
A plot of the mean percentage correctly identified facial emotions (taking together all participants) as a function of the 10 intensities at which the emotions in the FERT were displayed, showed a linear increase around the threshold of 50% accuracy. Yet, the individual pre- and post- intervention linear regressions, with % correct as dependent and emotion intensity as the independent variable, did not show optimal fit. All intensity coefficients (a) fitted (p < 0.05) significantly but the intercept values (b) were non-significant (p > 0.05) for all of the 128 ( = 64 participants x 2 sessions) linear regressions. Nevertheless, the individual 50% emotion identification threshold pre- and post-intervention appraised with the function (50 – b)/a were all intensities between 31.28% and 76.52% revealing no problematic
miscalculations. The variable “threshold” (i.e. the 50% emotion identification threshold) is normally distributed. Table 3 presents the mean 50% emotion identification threshold pre- and post-intervention for both groups (probiotics vs. placebo) in both pre-intervention emotion regulation categories (high MAIA vs. low MAIA).
Table 3.
Mean emotion intensity and standard deviation (shown in parentheses) of emotion expression at which 50% of the facial emotion was correctly identified for low (low MAIA) and high (high MAIA) ability to regulate emotion in the probiotics and the placebo groups. Asterisks indicate significant within group differences. .
Pre-intervention Post-intervention
Probiotics Low MAIA (N = 13) 49.57 (7.55) 46.53 (7.85)
High MAIA (N = 18) 53.97 (11.21) 47.39 (8.73) * Total (N = 31) 52.13 (9.94) 47.03 (8.25)
Placebo Low MAIA (N = 19) 51.61 (8.94) 46.50 (11.03) *
High MAIA (N = 14) 46.06 (9.48) 46.58 (13.85) Total (N = 33) 49.25 (9.45) 46.53 (12.09) * p < 0.05
Mixed ANOVA, with time (pre- and post-intervention) as within-subjects factor and both group (probiotics/placebo) and MAIA-categorization (low MAIA/high MAIA) as between-subject factors, revealed a significant main effect of time (pre- post intervention) on the 50% emotion identification threshold [F(1,60) = 12.101 , p < .001, ηp2 = .168]. Time by group interaction [F(1,60) = 1.515, p = .223, ηp2 = .025] and time by MAIA categorization interaction [F(1.60) = .262, p = .610, ηp2 = .004] were not found to be significant. Yet, the mixed ANOVA revealed a significant time by intervention group (placebo vs. probiotics) by MAIA-group (high MAIA vs. low MAIA) interaction F(1,60) = 5.029, p = .029, ηp2 = .077.
Bonferroni corrected pairwise comparison showed that there was a significant decrease of the 50% emotion identification threshold from pre [mean = 51,61, sd = 8.94] to post intervention [mean = 46.50, sd = 11.03, p = .008] for participants in the placebo intervention group with a low ability to regulate emotion at baseline. Also, there was a significant decrease of the 50% emotion identification threshold from pre-intervention [mean = 53.97, sd =11.21] to post-intervention [mean = 47.39, sd = 8.75, p = .001] for participants in the probiotics intervention group with a high ability to regulate emotion at baseline.
The aim of this study was to determine the impact of a 4-week multispecies probiotics intervention on facial emotion recognition, while taking a potential moderating effect of the ability to regulate emotion into account. Interoceptive awareness is positively related to emotion regulation ability (Remmers et al., 2016; Zamariola et al., 2019). Therefore, pre-intervention MAIA-scores were used to separate participants into two groups (high vs. low interoceptive awareness) before further analysis of probiotics intervention effects on 50% emotion identification threshold. The results show that the 50% emotion identification threshold significantly decreased for participants with a high ability to regulate emotion who received the probiotics intervention. As hypothesized, the 50% emotion identification
threshold did not significantly change for participants with a high ability to regulate emotion who received the placebo intervention (Figure 4). For participants with a low MAIA-score at baseline, the MAIA-score significantly increased from pre- to post-intervention. The results regarding the change of the 50% emotion identification threshold in this scenario, are not in line with our expectations as described in Figure 1. The 50% emotion identification threshold significantly decreased for participants with a low ability to regulate emotion who received the placebo intervention, but not for those who received probiotics (Figure 4). These results provide evidence that the ability to recognize facial emotion may be altered by manipulation of the gut microbiome with probiotics. Besides, a potential moderating effect of ability to regulate emotion is supported by the distinctive results found for two groups of participants with different ability to regulate emotion at baseline. It is important to mention that we can only speculate concerning the interpretation of the results discussed above, given the small sample size of the four participant groups and the resulting low power of our analysis.
Figure 4. Resulting effects of a 4-week (probiotics or placebo) intervention on ability to recognize facial emotion with ability to regulate emotion, represented by MAIA-score, as a moderating factor.
Decrease of the 50% emotion identification threshold for participants in the high pre-intervention MAIA-group, indicates that the emotion intensity at which 50% of the facial emotions was correctly identified, lowered. This result seems contradictory to earlier findings from Tillisch et al. (2013). They revealed reduced activity in brain regions processing emotion after a 4-week intervention with a fermented milk product containing probiotics. The brain regions they studied included primary interoceptive and somatosensory cortices, precuneus, frontal-, prefrontal- and temporal cortices, parahippocample gyrus and the periaqueductal gray [PAG]. A reduction in activity of this entire network after probiotics intervention was interpreted as dampening of emotional reactivity (Tillisch et al., 2013; Sarkar et al., 2018). Reduction of emotional reactivity contradicts an improved ability to recognize facial emotion when interpreted as increased sensitivity to facial emotion expression.
However, this increased sensitivity to facial emotion expression might actually be interpreted as an increased ability to correctly label presented facial emotion expression at a lower emotion intensity. Labelling presented emotional facial expressions goes beyond emotional reactivity per se. When facial expressions are presented, a subcortical route via the superior colliculus and the pulvinar thalamus activates the amygdala and other limbic system structures. The first responses of the human amygdala to emotional facial expressions are seen after 120 ms, even when the facial expressions are shown subliminal (Young et al. 2007). This brain activity represents the physiological emotional reactivity reported by Tillisch et al. (2013) and Sarkar et al. (2018). However, the activity of the amygdala is mediated by the frontal cortex. When subjects are asked to explicitly label emotions the amygdala response reduces, possibly by inhibiting activity from frontal cortices (Adolphs, 2002). Focus on connectivity within the brain network for the processing of emotion revealed a more dominant role for frontal cortices after probiotics intake (Tillisch et al., 2013). After the 4-week
probiotics intervention, PAG centered midbrain acivity was negatively correlated to activity of the occipital cortex, insula, cingulate cortex, basal ganglia, amygdala and somatosensory cortex (i.e. limbic system structures). Yet, it was positively correlated to activity of the prefrontal cortex (Tillisch et al., 2013). This was interpreted as a change from an arousal-based resting-state emotion processing network to a regulatory emotion processing network.
Implicating that probiotics reduce exhibition of mid- and forebrain structures, which reduces emotion reactivity (Tillisch et al., 2013; Steenbergen et al., 2015). At the same time, the activity of frontal structures in the brain increases, facilitating attention to emotional stimuli, as earlier reported by Bagga et al. (2018), and the ability to correctly label facial emotion expression as found in the current study.
Opposing the significant decrease of the 50% emotion identification threshold for participants in the high-MAIA group, no significant change of the 50% emotion identification threshold was found in the low-MAIA group after a 4-week probiotics intervention (Figure 4). This result is also contrary to the increase of the 50% emotion identification threshold that was expected. An increased 50% emotion identification threshold would have represented a “protective” function of probiotics against potential mental and/or physical harm. Increasing the emotion intensity for which 50% of the facial emotions are labelled correctly, is
interpreted as protective when talking about a group of participants who is potentially not able to successfully control their own emotion reactivity.
Nevertheless, the results do present that the effect of probiotics intervention on facial emotion recognition is not the same for participants with a different ability to regulate emotion at baseline. This supports the proposed moderating effect of emotion regulation on the impact of probiotics intervention on facial emotion recognition.
Completely unexpected, a significant lowering of the 50% emotion identification threshold was found after 4-weeks of placebo intake for participants with a low ability to regulate emotion at baseline (Figure 4). Initially, no change in ability to recognize facial emotion was expected to follow from intervention with placebo supplements. Hindsight, participants in the placebo condition might have underwent “intervention” despite taking the supplements without active substance. As explained in the procedure section, all participants were asked to fill out a short daily questionnaire in which they had to answer two (i.e.
positive- and negative) feelings questions. Daily reflecting on one’s own feelings might have caused changes in interoceptive awareness, emotion regulation and/or eventually emotion processing. Results regarding the pre- and post-intervention MAIA-scores indicate that interoceptive awareness did significantly increase for participants with a low MAIA-score at base line. According to the typological model of interoception from Garginel et al. (2015), interoceptive awareness (IA) is the relationship between interoceptive accuracy (IAcc) and interoceptive sensibility (IS). They identify interoceptive awareness as a measure of
correlation between objective ability to perceive bodily states (IAcc) and subjective interpretation of perceived bodily states (IS). The objective measures for interoceptive awareness (IAcc) never show change after body awareness-enhancing approaches such as daily reflecting on one’s own feelings (Mehling et al., 2012). This implicates that increased IA reflects improved subjective interpretation of perceived bodily states in both intervention groups as a result of daily reflecting on one’s own feelings. It is important to mention that the majority of interoceptive signals from the body, entering the brain in the brainstem (i.e. NTS), go unnoticed because they never reach consciousness. The afferent information that has not reached consciousness will not be reported in the MAIA, yet does impact emotional processes (Critchley & Harrison, 2013). Measurement of change in interoceptive awareness on a subconscious level is needed to elaborate on the role of interoception in emotion processing. If interoceptive awareness is as positively related to emotion reactivity as was implicated from earlier research (Jones, 1994; Montoya and Schandry, 1994; Cameron, 2001; Critchley et al, 2004), improved interoceptive awareness, as a result of daily reflecting on one’s own feelings, would explain the significant improvement of emotion recognition found for participants with low-MAIA scores at baseline despite receiving the placebo intervention.
However, participants who followed the 4-week probiotics intervention also reflected on their own feelings on a daily basis. The results even show that MAIA-scores increased for participants with a low MAIA-score at baseline regardless of the intervention condition. With the daily questionnaire proposed as “intervention” inducing this, you would expect the same for participants with a low MAIA-score in the probiotics condition. Considering that
probiotics supplementation also improved emotion recognition without significant change in MAIA-score for participants with a high MAIA-score at baseline, you might even expect a very strong improvement. Yet, the results show that the 50% emotion detection threshold did not change for participants with a low-MAIA score at base line who received a 4-week probiotics intervention. This non-effect is especially noteworthy as it does not only contradict our expectations, but also contradicts the proposed interpretation of results described thus far.
We want to propose the hypothesis that the ability to recognize facial emotion actually did improve for participants with a low ability to regulate emotion at baseline during the probiotics-intervention yet again decreased within the 4-week intervention period resulting in a non-effect comparing pre- and post-intervention. Improved emotion recognition parallel to an increase of interoceptive awareness, has probably lowered the 50% emotion detection
threshold. This was also implicated from the results for participants with low-MAIA score at baseline in the placebo condition described earlier. The intake of probiotics potentially
improved the ability to correctly label presented facial emotion expression at a lower emotion intensity like it did for participants with a high MAIA-score at baseline. However, the
proposed effect of probiotics intervention on top of the potential intervention effect of daily reflecting on feeling might have been too much resulting in a turnover. This reverses the effect and thus decreases the ability to recognize facial emotion. If daily reflecting on one’s own feelings indeed improves reactivity towards emotion stimuli as a result of increased interoceptive awareness. And probiotics improve the ability to correctly label facial emotion as a result of increased activity of frontal brain structure, self-protection is crucial.
Adaptations in affective, behavioural and cognitive mechanisms that underlie emotions (i.e. emotion processing) (Kret & Ploeger, 2015) might offer this protection. A change from an increasing to a decreasing ability to detect facial emotion, would protect against double improvement of sensitivity to emotion stimuli. Such an aggregated sensitivity may otherwise cause a problem in successfully regulating emotional reactivity. This could increase the risk of harming mental and/or physical health (Füstös et al., 2013; Kret & Ploeger, 2015). If emotion processing is indeed changing dynamically over time to offer protection against mental harm, this may be an important aspect to focus on in treatment of psychological diseases like depressions. Common contradictory and seemingly non-existent effects of anti-depressives on emotion processing (review: Merens et al., 2007) may also be explained by the proposed turnover effect or either by a disruption of this potentially protective mechanism. Further research should, first of all, focus on applicable research designs to test whether the proposed turnover effect might happen over the course of intervention between pre-and post-intervention measures. Furthermore, the focus should be on possible adaptations in emotion processing offering the protection underlying the proposed turnover effect.
The current study has a few limitations that have to be mentioned. First, despite exclusion of participants who took probiotics supplements in the past 3 months, we did not control for intake of probiotics rich food (e.g. fermented products) during the intervention. Hence dietary choices may have indirectly contributed to the effects that we found. Second, females taking hormonal contraceptives were not excluded from the study and the female participants in this study were not controlled for their menstrual cycle. With an intervention of exactly 4 weeks the influence of the menstrual cycle on emotion processing (Sundström & Gingnell, 2014) was assumed to be the same during the pre- and post-intervention session. Yet multiple
studies have revealed impact of oral contraceptives on core social-emotional behaviours and brain systems (review: Montoya & Bos, 2017) potentially influencing emotional processing as measured in the current study. Third, in this study the Multidimensional Assessment of Interoceptive Awareness (MAIA; Mehling et al., 2012), a self-report measure for interoceptive body awareness, has been used as an indication for the ability to regulate emotion based on the positive association between interoception and emotion regulation (Critchley & Harrison, 2013; Füstös et al., 2013; Price & Hooven, 2018; Zamariola et al., 2019). This could be improved by using a more direct measure for emotion regulation ability, such as the ‘Emotion Regulation of Others and Self’ [EROS] scale (Niven et al., 2011). Finally, the 50% emotion identification threshold has been determined assuming a linear positive association between displayed emotion intensity and accuracy in the FERT. Even though, the fit of the pre- and post-intervention linear regression, with % correct as dependent and emotion intensity as the independent variable, was non-optimal. Analysis of plots with accuracy as a function of emotion intensity for all of the displayed emotions individually reveal an s-shaped sigmoid-curve and not a linear association. Investing in a method to extrapolate the 50% emotion identification threshold from a sigmoid-curve would result in more accurate measures for facial emotion recognition.
Future, research on the impact of probiotics intervention on psychosomatic processes should be conducted more widely on emotion regulation. In this study assessment of
interoceptive awareness was used as a measure for emotion regulation ability. Emotion regulation controls the offset of emotional responding by targeting emotion-generating systems (Koole et al., 2009). Physiological changes are acknowledged as an important contribution to emotion generation (Gendron and Barrett, 2009). Interoception may therefore be a target point for emotion regulation. However, what direction of change in interoceptive awareness would represent improved emotion regulation, may be debated. Improved
conscious awareness of interoceptive information contributes to the ability to understand and recognize emotion which provokes regulation of emotion (Zamaroila et al., 2019). Yet, a core feature of emotion regulation is activation of a goal. The common goal is down-regulation of negative emotions and up-regulation of positive emotions. However, other emotion regulation goals are also possible (Gross et al., 2014). Increased interoceptive awareness as a
representation of an improved ability to regulate emotion is not informative for the preferred direction of regulation and the success of regulating in that direction. Focusing on the use of probiotics as an antidepressant, future research should study the effects of probiotics on
emotion regulation specifically beneficial for depressed patients. However, since the effect of acknowledged antidepressants on emotion reactivity is still inconclusive (Merens et al., 2007), the “antidepressive effect” regarding emotion reactivity has to be revealed first. To do so we should not only add more objective measurement, but also pay attention to subjective reports. Eventually the problem in depression is not biological nor even a change in cognitive or affective processing per se. It rather is an individual reporting that s/he feels depressed. The ultimate “anti-depressive effect” is a situation in which an individual no longer reports this subjective depressed feeling. When the emotion regulation goal, resulting in an objectively and/or subjectively reported antidepressant effect, is set, research can elaborate on whether probiotics promote emotion regulation in this specific direction.
To conclude, this study revealed that the ability to recognize facial emotion may be altered by manipulation of the gut microbiome with a probiotics intervention. It was found that a probiotics intervention increased the ability to identify facial emotion specifically for people with a high ability to regulate emotion at baseline. Dynamic changes in
emotion processing were proposed as an explanation for aberrant results for people with a low ability to regulate emotion. These findings contribute to the extension of thus far scarce research on the impact of microbiota manipulation on affective processes. Though, the low power of our analyses stresses that these results should be seen as preliminary. To learn more about the impact of probiotics intervention on emotion processing and
eventually study the implications of those results for probiotics as potential treatment for mental disorders, further investigation is of necessity.
References
Abid, M. B., & Koh, C. J. (2019). Probiotics in health and disease: fooling Mother Nature? Infection, 1-7.
Adolphs, R. (2002). Neural systems for recognizing emotion. Current opinion in neurobiology, 12(2), 169-177.
Bagga, D., Reichert, J. L., Koschutnig, K., Aigner, C. S., Holzer, P., Koskinen, K., Moissl-Eichinger, C., & Schöpf, V. (2018). Probiotics drive gut microbiome triggering emotional brain signatures. Gut Microbes, 9(6), 486-496.
Baron-Cohen, S., & Wheelwright, S. (2004). The empathy quotient: an investigation of adults with Asperger syndrome or high functioning autism, and normal sex
differences. Journal of autism and developmental disorders, 34(2), 163-175.
Bastiaanssen, T. F., Cussotto, S., Claesson, M. J., Clarke, G., Dinan, T. G., & Cryan, J. F. (2020). Gutted! Unraveling the Role of the Microbiome in Major Depressive
Disorder. Harvard Review of Psychiatry, 28(1), 26-39.
Bravo, J. A., Forsythe, P., Chew, M. V., Escaravage, E., Savignac, H. M., Dinan, T. G., Bienenstock, J., & Cryan, J. F. (2011). Ingestion of Lactobacillus strain regulates emotional behavior and central GABA receptor expression in a mouse via the vagus nerve. Proceedings of the National Academy of Sciences, 108(38), 16050-16055.
Beck, A.T. (1967). Depression: Causes and Treatment. University of Pennsylvania Press, Philadelphia.
Benton, D., Williams, C., & Brown, A. (2007). Impact of consuming a milk drink containing a probiotic on mood and cognition. European journal of clinical nutrition, 61(3), 355-361.
Bonaz, B., Bazin, T., & Pellissier, S. (2018). The vagus nerve at the interface of the microbiota-gut-brain axis. Frontiers in neuroscience, 12, 49.
Bomfim, A. J. D. L., Ribeiro, R. A. D. S., & Chagas, M. H. N. (2019). Recognition of dynamic and static facial expressions of emotion among older adults with major
depression. Trends in psychiatry and psychotherapy, 41(2), 159-166.
Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of personality and social psychology, 63(3), 452.
Calder, A. J., Young, A. W., Perrett, D. I., Etcoff, N. L., & Rowland, D. (1996). Categorical perception of morphed facial expressions. Visual Cognition, 3(2), 81-118.
Cameron, O. G. (2001). Interoception: the inside story—a model for psychosomatic processes. Psychosomatic medicine, 63(5), 697-710.
Cenit, M. C., Sanz, Y., & Codoñer-Franch, P. (2017). Influence of gut microbiota on neuropsychiatric disorders. World journal of gastroenterology, 23(30), 5486.
Claesson, M. J., Jeffery, I. B., Conde, S., Power, S. E., O’Connor, E. M., Cusack, S., Harris, H.M.B., Coakley, M., Lakshminarayanan, B., O’Sullivan, O., Fitzgerald, G.F., Deane, J., O’Connor, M., Harnedy, N., O’Connor, K., O’Mahony, D., Sinderen, D., Wallace, M., Brennan, L., Stanton, C., Marchesi, J.R., Fitzgerald, P.A., Shanahan, F., Hill, C., Ross, R.P., & O’Toole, P.W. (2012). Gut microbiota composition correlates with diet and health in the elderly. Nature, 488(7410), 178-184.
Cohen, S., Kamarck, T., & Mermelstein, R. (1994). Perceived stress scale. Measuring stress: A guide for health and social scientists, 10.
Coupland, N. J., Singh, A. J., Sustrik, R. A., Ting, P., & Blair, R. J. (2003). Effects of diazepam on facial emotion recognition. Journal of Psychiatry and Neuroscience, 28(6), 452.
Craig, A. D. (2002). How do you feel? Interoception: the sense of the physiological condition of the body. Nature reviews neuroscience, 3(8), 655-666.
Critchley, H. D., & Harrison, N. A. (2013). Visceral influences on brain and behavior. Neuron, 77(4), 624-638.
Critchley, H. D., Wiens, S., Rotshtein, P., Öhman, A., & Dolan, R. J. (2004). Neural systems supporting interoceptive awareness. Nature neuroscience, 7(2), 189-195.
Cryan, J. F., & Dinan, T. G. (2012). Mind-altering microorganisms: the impact of the gut microbiota on brain and behaviour. Nature reviews neuroscience, 13(10), 701-712.
Dallal, G. E. (2007). randomization.com. Retrieved May 20, 2020, from http://randomization.com/
Derrien, M., & van Hylckama Vlieg, J. E. (2015). Fate, activity, and impact of ingested bacteria within the human gut microbiota. Trends in microbiology, 23(6), 354-366.
van der Does, A. J. W., & Williams, J. M. G. (2003). Leiden index of depression sensitivity–revised (LEIDS-R). Leiden University.
Ekman, P., & Friesen W.V. (1976). Pictures of facial affect. Consulting Psychologists Press.
Everaert, J., Duyck, W., & Koster, E. H. (2014). Attention, interpretation, and memory biases in subclinical depression: A proof-of-principle test of the combined cognitive biases hypothesis. Emotion, 14(2), 331.
Fan, J., McCandliss, B. D., Sommer, T., Raz, A., & Posner, M. I. (2002). Testing the efficiency and independence of attentional networks. Journal of cognitive
neuroscience, 14(3), 340-347.
FAO/WHO. (2001). Evaluation of health and nutritional properties of powder milk and live lactic acid bacteria. Report from FAO/WHO Expert Consultation, 1-4.
Frank, M.G. (2001). Facial Expressions. International Encyclopedia of the Social & Behavioral sciences, 5230-5234.
Füstös, J., Gramann, K., Herbert, B. M., & Pollatos, O. (2013). On the embodiment of emotion regulation: interoceptive awareness facilitates reappraisal. Social cognitive and affective neuroscience, 8(8), 911-917.
Gaykema, R. P., Goehler, L. E., & Lyte, M. (2004). Brain response to cecal infection with Campylobacter jejuni: analysis with Fos immunohistochemistry. Brain, behavior, and immunity, 18(3), 238-245.
Gendron, M., & Feldman Barrett, L. (2009). Reconstructing the past: A century of ideas about emotion in psychology. Emotion review, 1(4), 316-339.
de Groot, A. C., & Steenbergen, L. (2019). Antibiotic use associated to decreased recognition of facial emotion expressions and increased aggressive cognitive reactivity: a preliminary study.
Harmer, C. J., Bhagwagar, Z., Perrett, D. I., Völlm, B. A., Cowen, P. J., & Goodwin, G. M. (2003b). Acute SSRI administration affects the processing of social cues in healthy volunteers. Neuropsychopharmacology, 28(1), 148-152.
Harmer, C. J., Heinzen, J., O’Sullivan, U., Ayres, R. A., & Cowen, P. J. (2008). Dissociable effects of acute antidepressant drug administration on subjective and emotional processing measures in healthy volunteers. Psychopharmacology, 199(4), 495-502.
Harmer, C. J., Hill, S. A., Taylor, M. J., Cowen, P. J., & Goodwin, G. M. (2003a). Toward a neuropsychological theory of antidepressant drug action: increase in positive emotional bias after potentiation of norepinephrine activity. American Journal of Psychiatry, 160(5), 990-992.
Harmer, C. J., O’Sullivan, U., Favaron, E., Massey-Chase, R., Ayres, R., Reinecke, A., Goodwin, G.M., & Cowen, P. J. (2009). Effect of acute antidepressant administration on negative affective bias in depressed patients. American Journal of Psychiatry, 166(10), 1178-1184.
Harmer, C. J., Shelley, N. C., Cowen, P. J., & Goodwin, G. M. (2004). Increased positive versus negative affective perception and memory in healthy volunteers following selective serotonin and norepinephrine reuptake inhibition. American Journal of
Psychiatry, 161(7), 1256-1263.
Horder, J., Cowen, P. J., Di Simplicio, M., Browning, M., & Harmer, C. J. (2009). Acute administration of the cannabinoid CB1 antagonist rimonabant impairs positive affective memory in healthy volunteers. Psychopharmacology, 205(1), 85-91.
Ingram, R. E., Miranda, J., & Segal, Z. (2006). Cognitive vulnerability to depression. In Cognitive vulnerability to emotional disorders (pp. 73-102). Routledge.
Jones, G. E. (1994). Perception of visceral sensations: A review of recent findings, methodologies, and future directions.
Kelly, J. R., Allen, A. P., Temko, A., Hutch, W., Kennedy, P. J., Farid, N., Murphy, E., Boylan, G., Bienenstock, J., Cryan, J.F., Clarke, G., & Dinan, T.G. (2017). Lost in translation? The potential psychobiotic Lactobacillus rhamnosus (JB-1) fails to modulate stress or cognitive performance in healthy male subjects. Brain, behavior, and immunity, 61, 50-59.
Kemmis, L., Hall, J. K., Kingston, R., & Morgan, M. J. (2007). Impaired fear recognition in regular recreational cocaine users. Psychopharmacology, 194(2), 151-159.
Koole, S. L. (2009). The psychology of emotion regulation: An integrative review. Cognition and emotion, 23(1), 4-41.
Kret, M. E., & Ploeger, A. (2015). Emotion processing deficits: a liability spectrum providing insight into comorbidity of mental disorders. Neuroscience & Biobehavioral Reviews, 52, 153-171.
Lawless, H. T. (2010). A simple alternative analysis for threshold data determined by ascending forced‐choice methods of limits. Journal of Sensory Studies, 25(3), 332-346. Lewis, S. J., & Heaton, K. W. (1997). Stool form scale as a useful guide to intestinal transit time. Scandinavian journal of gastroenterology, 32(9), 920-924.
Lozupone, C. A., Stombaugh, J. I., Gordon, J. I., Jansson, J. K., & Knight, R. (2012). Diversity, stability and resilience of the human gut microbiota. Nature, 489(7415), 220-230.
Mallick-Searle T. The microbiome: pain and mood. Presented at: 2019 American Association of Nurse Practitioners (AANP) Annual Meeting; June 18-23; Indianapolis, IN.
Mayer, E. A., Knight, R., Mazmanian, S. K., Cryan, J. F., & Tillisch, K. (2014). Gut microbes and the brain: paradigm shift in neuroscience. Journal of Neuroscience, 34(46), 15490-15496.
