Mindfulness is commonly defined as the capacity, or ability of individuals to be conscious of and attend internal and external events in an open discerning way, without judging these experiences (Shapiro and Carlson, 2009). Mindfulness has been associated with improved cognitive performance, including inhibitory control and attention (Zoogman et al., 2015; Gallant, 2016; Klingbeil et al., 2017; Quaglia et al., 2019; Vekety et al., 2021). However, the moderating role of reward context is not yet fully understood, which is of importance to disorders of addiction. Specifically, addiction is in part characterized by attentional bias for reward-associated stimuli which contributes to associated maladaptive behavior (Robinson et al., 2016; Volkow et al., 2017). However, studies on the effects of mindfulness regarding attentional bias in a reward context are relatively mixed. Some studies suggest that attentional bias is reduced following mindfulness interventions (Garland et al., 2012; Alamout et al., 2020), and other studies suggest enhanced attentional bias for reward-associated stimuli (Delgado-Pastor et al., 2013; Sanger et al., 2018; Isbel et al., 2019; Logemann-Molnár et al., 2022). The aim of the current work was to address this apparent discrepancy in a controlled lab environment.

The neuroanatomical correlates of mindfulness and their association with executive functions are not yet fully understood, but studies suggest that the brain mechanism overlaps between dispositional mindfulness and mindfulness training-induced mindfulness (Bilevicius et al., 2017; Feruglio et al., 2021). Previous studies incorporating brain activity measures have shown that mindfulness is associated with suppressed activity within the Default Mode Network, which is inversely associated with activity in the Salience Network (Bilevicius et al., 2017; Feruglio et al., 2021). This may translate to mindfulness-associated enhanced susceptibility for salient stimuli and subsequent approach tendencies (Bilevicius et al., 2017; Feruglio et al., 2021). This effect is further supported by studies that show that mindfulness is associated with more left relative to right frontal brain activity, indexed by frontal alpha asymmetry using electroencephalography (EEG) (Keune et al., 2013; Moynihan et al., 2013), which has also been associated with enhanced approach tendencies (Kelley et al., 2017). However, it should be noted that other studies did not find such effect (Keune et al., 2011; Szumska et al., 2021), and the effect may be state dependent (Keune et al., 2013).

Based on the above, mindfulness may be associated with enhanced attentional bias for salient stimuli. Behavioral studies support this notion. Studies that implemented an oddball paradigm show mindfulness-associated attentional capture of deviant infrequent, but salient stimuli in the context of repetitive stimuli (Delgado-Pastor et al., 2013; Sanger et al., 2018; Isbel et al., 2019). ¹³²⁴²⁵ However, studies on addiction that employed variants of a visual probe task and have focused on the association between mindfulness and attentional bias for reward related stimuli, may seem incongruent at first glance. Typically, in these studies, mindfulness reduced attentional bias for reward-associated stimuli (Garland et al., 2012; Alamout et al., 2020). One plausible explanation for this apparent discrepancy may be related to differences in the expectedness of the salient, reward-related stimulus. It should be noted that visuospatial attention consists of two main components, voluntary (top-down) attentional orienting, and stimulus-driven (bottom-up) reflexive attentional (re-)orienting (Corbetta and Shulman, 2002; Corbetta et al., 2008). Voluntary attention allows for the conscious focus of attention to potentially relevant stimuli in our environment, whereas stimulus-driven attentional (re-)orienting allows for attentional disengagement and subsequent shift towards unanticipated, yet relevant, salient stimuli (Corbetta and Shulman, 2002; Corbetta et al., 2008). It seems plausible that mindfulness exerts differential effects on these components. To elaborate, reward-associated stimuli in oddball paradigms are commonly infrequently presented and relatively unexpected. In such case, results seem to reflect that mindfulness is associated with enhanced stimulus-driven attentional capture of such stimuli. In contrast, reward-associated stimuli in visual probe paradigms are expected and frequently presented. In such studies, mindfulness is associated with reduced voluntary attention towards such stimuli.

One excellent paradigm to investigate and differentiate voluntary attention from stimulus-driven attention to unexpected relevant stimuli is the Posner paradigm, also called the visuospatial cueing task (Posner et al., 1980). In such task, a cue indicates the likely location of a subsequent target to which a response is required. In a minority of trials, the target is presented at a location opposite as indicated by the cue. Generally, individuals respond faster to targets that are validly cued, and thus presented in the attended visual hemifield as opposed to targets that are invalidly cued and presented in the unattended visual hemifield. The relevant index is the validity effect, which is operationalized as the difference in response time to validly relative to invalidly cued targets (Posner et al., 1980). The validity effect represents the benefit in terms of speeded response times due to valid cueing relative to response time costs in the context of invalid cueing. It can be argued that enhanced voluntary attention in response to valid cueing results in faster responses and an enhanced validity effect. In contrast, stronger stimulus-driven attention is associated with enhanced attentional disengagement and reorienting in the context of invalid cueing, and results in a reduction of the validity effect (Corbetta and Shulman, 2002; Corbetta et al., 2008). In the VSC task, reward context can be operationalized by target characteristics (Failing and Theeuwes, 2014; Tsegaye et al., 2022). Specifically, an intrinsic reward context can be operationalized by target stimuli representing palatable food, which is known to trigger reward circuity in the brain, similarly to monetary target stimuli (Yousuf et al., 2018). Indeed, it may be questioned whether this will also translate into observable effects in normal healthy samples. In that vein, it should be noted that a recent systematic review showed no consistent difference in attentional bias comparing overweight/obese individuals to healthy controls (Hagan et al., 2020). This is also supported by our previous online study that suggested enhanced attentional bias in a palatable food relative to a neutral context, irrespective of BMI (Tsegaye et al., 2020).

Integrating the results from the aforementioned studies, the following hypotheses were postulated. Firstly, it was hypothesized that mindfulness would be associated with a reduced validity effect, reflecting a combination of reduced voluntary attention and increased stimulus-driven attention. Secondly, it was expected that this effect would be higher in a reward context, operationalized by palatable food associated stimuli, relative to the neutral context. Lastly, we explored the relationship between mindfulness and frontal alpha asymmetry as a brain activity index of approach tendency.

For the main VSC task analyses, participants (N = 72) were included that completed the VSC task and were engaged with the task, indicated by performance above chance level during all conditions and trial-types (proportion correct >0.5). We should note that the main results of the selected sample did not differ in terms of significance from the explorative whole group analyses. The MAAS score ranged from 2.40 to 5.33 (M = 3.93, SD = 0.60). Dispositional mindfulness did not affect the relationship between reward context (neutral/food) and validity (valid/neutral/invalid) regarding response time, F(1.79,125.41) = 0.30, p = 0.715, η_p² = 0.004. Hence, this does not support that the relationship between dispositional mindfulness and voluntary relative to stimulus-driven driven attention differs between the conditions. However, as evident in Figure 2, dispositional mindfulness significantly reduced voluntary attention relative to stimulus-driven attention irrespective of reward context, as indicated by a significant interaction between dispositional mindfulness and validity (valid/neutral/invalid) regarding response time, F(1.77,124.02) = 3.73, p = 0.032, η_p² = 0.051. This was also confirmed by subsequent post-hoc analysis, which revealed a reduced response time difference between validly cued targets and invalidly cued targets as a function of dispositional mindfulness level, F(1,70) = 4.70, p = 0.034, η_p² = 0.063. There was an overall effect of cuing on response time. Specifically, the main effect of validity (valid/neutral/invalid) regarding response time was significant, F(1.77,124.02) = 6.75, p = 0.002, η_p² = 0.088. Post-hoc testing revealed that responses were significantly faster for validly cued targets relative to invalidly cued targets [F(1,70) = 8.51, p = 0.005, η_p² = 0.108], underscoring the validity of the paradigm. There was no main effect of dispositional mindfulness regarding response time, F(1, 70) = 3.04, p = 0.086, η_p² = 0.042. Lastly, There was no significant association between dispositional mindfulness and frontal alpha asymmetry in both the eyes closed and eyes open condition, respectively r(95) = 0.034, p = 0.741, and r(95) = −0.084, p = 0.417. The significance did not change after the exclusion of outliers (defined as exceeding 2 SD from the mean).

The current study focused on the relationship between dispositional mindfulness and voluntary and stimulus driven attention in contexts that differed in terms of intrinsic reward. The results partly confirmed our hypotheses. Specifically, dispositional mindfulness was associated with a reduction of the validity effect, plausibly reflecting a combination of reduced voluntary attention and enhanced stimulus-driven attention. In contrast to our hypotheses, this effect did not differ as a function of reward context. In addition, the observed relationship between dispositional mindfulness and visuospatial attention could not be explained by asymmetry of frontal brain activity.

Results of the current study suggest that mindfulness differentially impacts voluntary and stimulus-driven attention as indicated by the reduced validity effect. Interestingly, we did not find that reward context moderated the effect of dispositional mindfulness on the validity effect. It might be argued that such effect might be restricted to individuals affected by addiction and obesity (noting the considerable overlap of the latter with addiction in terms of behavioural challenges as well as the associated brain mechanism; Tsegaye et al., 2020). On the other hand, palatable food stimuli (intrinsically rewarding stimuli), have been shown to induce attentional bias in healthy participants, not only in obesity (Hagan et al., 2020). Either way, our sample consisted of healthy participants, and the paradigm included a narrow set of palatable food stimuli. Hence, we cannot fully exclude the possibility that effects might be different in addiction using a different set of stimuli associated with the drug of choice.

We also explored the potential relationship between dispositional mindfulness and asymmetry of frontal brain activity, indexed by FAA. If such relationship would be evident, FAA could be an interesting target for neuromodulatory approaches such as Transcranial Alternating Current Stimulation, Unilateral Muscle Contraction, or perhaps EEG-feedback, to potentially supplement mindfulness training (Kelley et al., 2017). However, results did not show evidence of such relationship. Though this could indicate that the relationship between dispositional mindfulness and visuospatial attention cannot be explained by asymmetry of frontal brain activity, appropriate nuance should be applied. It should be underscored that we did not perform a mindfulness intervention but assessed dispositional mindfulness and did not select individuals based on high vs. low mindfulness, which could be considered a limitation. As such, the range of dispositional mindfulness levels might have been too limited to detect an effect on FAA. On the other hand, it remains a possibility that the relationship between dispositional mindfulness and visuospatial attention is best explained by a different electrophysiological mechanism.

Some other limitations of the current study should also be noted. Based on previous studies that suggested a moderating role of reward context regarding the relationship between subject characteristics and executive control (Houben et al., 2014; Tsegaye et al., 2020), we included depictions of both savory and sweet snacks as target stimuli for the food context. However, we did not control for individual differences in food preference, which may result in enhanced variability and associated reduced statistical power to detect a potential moderating effect. Also, we must be careful with generalizing across other reward contexts. As argued before, effects may vary as a function of the exact operationalization of reward context.

Lastly, it may be argued that our implementation of the reward context does not only differ in terms of reward from the neutral context, but also in terms of stimulus complexity. It has been shown that higher stimulus complexity results in increased response times, due to enhanced processing that is required for such stimuli (Gajewski and Falkenstein, 2013). However, our data does not show a significant effect of reward context on mean response time (collapsed across trial-type), F(1,70) = 2.20, p = 0.143, η_p² = 0.030.

In conclusion, together with previous observations, our results imply that mindfulness reduces voluntary goal-directed (top-down) attention while enhancing stimulus-driven (bottom-up) attention to salient unexpected stimuli in our environment.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: Open Science Framework public repository: DOI 10.17605/OSF.IO/4965P.

The studies involving humans were approved by Research Ethics Committee of the Institute of Psychology, Eötvös Loránd University (ELTE). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

ZL-M: Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing. AV-S: Conceptualization, Funding acquisition, Methodology, Writing – review & editing. ZD: Conceptualization, Funding acquisition, Methodology, Resources, Writing – review & editing. HL: Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Software, Visualization, Writing – review & editing.