Pituitary adenoma (PA) is one of the common benign intracranial tumors of the central nervous system, accounting for almost 15% of all cases (Hauser et al., 2019; Melmed, 2020). Emerging studies have established cognitive impairments in executive control, attention, and working memory both before and after surgery in patients with pituitary adenoma (Psaras et al., 2011; Butterbrod et al., 2019; Pertichetti et al., 2020), but the underlying neurophysiological mechanisms remain unclear. The physical compression, treatment strategy, surgical approach, and especially abnormal hormone levels may be responsible for the impairment of cognitive function (Peace et al., 1997, 1998; de Oliveira et al., 2008; Tooze et al., 2009; Tooze and Sheehan, 2018). In Yao’s opinion (Yao et al., 2017), prolactinoma patients exhibited decreased gray matter volume (GMV) in the left hippocampus, left orbitofrontal cortex, right middle frontal cortex, and right inferior frontal cortex, providing seminal evidence for deficits in verbal memory and executive control in patients with prolactinoma. The dysfunction of attention and inhibition control has been extensively examined in patients with pituitary adenomas based on neuropsychological scales and behavioral outcomes (Müssig et al., 2011; Pertichetti et al., 2020). To our knowledge, however, there have been few studies investigating the lateralized attentional networks in PA patients systematically.

According to the classical attention network theory, proposed by Posner and colleagues (Posner and Petersen, 1990; Petersen and Posner, 2012; Posner et al., 2019), the attention network has been divided into alerting, orienting, and executive control networks: Alerting refers to the state of obtaining and maintaining vigilance to upcoming information. This network may be associated with the right hemisphere (RH), frontal, parietal, and thalamic regions, and influenced by the norepinephrinergic system. Orienting network is for selecting specific information from the environment by focusing on one modality or location. Orienting is related to the cholinergic system and is associated with the frontal eye field (FEF), intraparietal sulcus (IPS), and other areas. Executive control network plays a major role in monitoring and resolving task-related conflicts, including error decision-making, and planning. This network primarily involved the dorsal anterior cingulate cortex and the lateral prefrontal cortex and corresponded to the dopaminergic system.

Despite their bilateral distribution, studies have shown that attentional functions might be dominated in the right hemisphere, particularly in the right parietal area (Mesulam, 1999; Brooks et al., 2014). Furthermore, for visuospatial attention, attending to one side of the visual stimuli typically corresponds to the activation of contralateral parietal areas. Based on the above findings and the ANT paradigm, Greene and colleagues proposed the LANT paradigm to assess the hemispheric asymmetry in each attention network as well as the attentional volume of each hemisphere (Greene et al., 2008). The LANT paradigm revealed the hemispheric lateralization of each attention network by rotating the original up and down target stimuli by 90° and presenting to the left and right visual fields (LVF, RVF). Previous studies of attentional networks indicated that multiple disorders, including mild traumatic brain injury (Chen et al., 2021) and attention deficit hyperactivity disorder (ADHD) (Adólfsdóttir et al., 2008; Lundervold et al., 2011), have the potential to impair attentional networks. Apart from a few studies related to stroke (Russell-Giller et al., 2021) and cerebral small vessel disease (Cao et al., 2020), to the best of our knowledge, systematic and comprehensive investigations based on the theory of the lateralized attention network in patients with pituitary adenoma were limited.

Referring to the comparable Attention Network Test (ANT) studies (Fan et al., 2002, 2005; Neuhaus et al., 2010; Williams et al., 2016), the cue-N1 component was associated with alerting and orienting networks. N1 was defined as an early visual attention component that appeared 150–250 ms after the cue stimulus and was distributed in the parietal and occipital regions. N1 was considered an early visual processing of stimulus properties, and the amplitude increased when the visual stimulus appeared in the attended spatial location. Furthermore, N1 also reflected the facilitation of early preattentive processing (Kaufman et al., 2016). Target-P3 components were associated with the executive control network and typically appeared around 250-500 ms after target presentation. P3 originated in the anterior cingulate gyrus and was located in the central parietal region, which mirrored the response inhibition process and attentional resource allocation (Polich, 2004, 2007).

In summary, in the present study, the LANT paradigm and ERPs were combined to investigate the hemispheric lateralization of attention networks in patients with pituitary adenoma. We hypothesized that (1) PA patients have significantly decreased behavioral and ERP results in attention networks; (2) Attention networks might be exhibited hemispheric asymmetry in two groups; (3) The pattern of hemispheric asymmetry in PA patients might be different from that in HCs; and (4) The serum prolactin (PRL) level may impair attention networks in patients with pituitary adenoma.

We combined a lateralized attention network paradigm with event-related potential techniques to examine the impairment and lateralization of each attention subnetwork in pituitary adenoma patients. In terms of behavioral performance, the temporal and spatial cue effects, as well as the flanker effect, could be evoked significantly by the LANT task. In line with previous findings based on ANT and LANT paradigms (Greene et al., 2008; Neuhaus et al., 2010; Thiebaut de Schotten et al., 2011), the decreased reaction time reflected the facilitatory effect of temporal and spatial cues on target responses. Both groups exhibited similar error rates in each subnetwork, but the PA group showed prolonged RTs in all conditions relative to HCs. To maintain relatively high accuracy for the same cue and target stimuli, excessed attentional resources have to pay for conflict resolution. Thus, we believed that the pituitary adenoma was more likely to impair global attentional processing, which has been demonstrated in previous studies (Chen et al., 2022).

Network efficiency results revealed that the PA group had lower alerting network efficiency but higher executive control network efficiency relative to the HCs. Given the differences in the visual field, a significant LVF alerting effect was found in the PA group. Although some studies have shown a significant LVF-RH dominance for alerting (Heilman and Van Den Abell, 1979; Funnell et al., 2003), these warning cues were only presented on the unilateral visual field instead of bilateral visual fields. In line with the present study, the results of Asanowicz (Asanowicz et al., 2012) indicated no visual field differences in the alerting network efficiency, which was induced by a revised LANT paradigm. Therefore, the visual field differences in alerting network in the PA group could be explained by the small sample size and the heterogeneity of tumor types. Regarding the inconsistent alerting network results in the N1, we hypothesized that RT was an indicator that measured alerting indirectly, Because RT reflected the full stages of cognition from receiving warning signals, arousing the alerting state to making a response ultimately. ERPs, on the other hand, can capture and quantify the alerting state specifically, and may not be predicted accurately by the behavioral results.

Although greater executive control efficiency was found in the PA group compared to HCs, further results from the ANOVA revealed that the PA group had prolonged RTs in both flanker conditions, and the difference was still greater than the HCs. Whereas the PA group responded slower in both target conditions, and significantly longer RTs were observed in the PA group when an incongruent condition was presented. Thus, we believed that the smaller efficiency of the executive control network indicated greater executive functions, which was consistent with previous evidence explained by the right hemisphere specialization in executive functions (Milham et al., 2001; Asanowicz et al., 2012). Overall, we hypothesized that greater efficiency of the executive control network corresponded to worse executive functions, indicating dysfunction of inhibition control in the PA group.

P3 modulated the inhibition control processes to the target stimuli, as well as the allocation of attentional resources (Polich, 2007; Kratz et al., 2011). In the HCs, the direction of P3 evoked by the target was opposite to RT, with larger P3 amplitudes corresponding to shorter RTs, which may be interpreted as task difficulty (Polich, 1987). More difficult tasks required greater attentional demands and evoked smaller P3 components, in other words, decreased P3 components mirrored the limitation of attentional capacity. Therefore, reduced P3 amplitudes across conditions in the PA group were indicative of executive control dysfunction and decreased attention allocation, which was supported by previous findings in PA patients (Tooze et al., 2009). The LANT task provided a notable advantage for investigating the lateralization of attention networks. The stimuli in the ANT task were presented on the midline, thus no visual field differences were induced, whereas the stimuli in the LANT task were presented in both visual fields and might be induced by the hemispheric differences in the P3 component. Studies have found a right hemisphere dominance in executive functions, whereas the left hemisphere has been shown to play an important role in semantic information processing (MacLeod, 1991; Russell-Giller et al., 2021). Therefore, the RCP distribution of the P3 component in the present study supported the RH dominance theory of executive function. Furthermore, the ANOVA results also exhibited a significant VF × hemisphere interaction, which indicated that the RH dominated the bilateral visual field information, while the LH only predominantly processed the left visual field information. The interaction results partially supported the theory of hemispheric lateralization (Nobre et al., 1997). Indeed, studies with a rapid serial visual presentation (RSVP) task have demonstrated that larger RH P3 was evoked in the RVF, whereas the LVF predominantly evoked the P3 in the LH, indicating a significant ipsilateral activation (Zhang et al., 2022). Further efforts at lateralization were needed to interpret the potential VF asymmetry of the executive control network.

Surprisingly, the 4-way (Flanker×VF × Hemisphere×Group) ANOVA ERPs results showed a main group effect in the pattern of hemispheric asymmetry when the flanker factor was included. Unlike the HCs group with a right hemispheric dominance when an incongruent target presented on the RVF, the PA group manifested a bilateral distribution of the P3. One reasonable interpretation was that reduced RCP capacity to RVF stimuli led to a compensatory activation in the LCP, highlighting the recruitment of resources from the non-task-dominant hemisphere. This view was consistent with Paitel’s findings for age-related alteration in inhibitory control, whereas compensatory recruitment occurred at low-moderate task demands, which was indicative of depleted neural reserves (Paitel and Nielson, 2021). However, different from the stop-signal task (SST), the P3 component evoked by LANT was correlated with the allocation of attentional resources and conflict resolution. Therefore, under high attentional demand, additional contralateral hemispheric resources of attention were recruited for task response, leading to enhanced activation in the LCP and the absence of RH dominance. In Weissman and Welcome’s view (Weissman and Banich, 2000; Welcome and Chiarello, 2008), increasing task difficulty required the recruitment of both hemispheres, resulting in an attenuated hemispheric asymmetry. Overall, to overcome the decreased attentional function in the RCP under high conflict conditions, attentional resources in the LCP were required in patients with pituitary adenomas, which led to the elimination of the RH-dominated pattern.

This pattern of contralateral activation in the orienting network was consistent with previous evidence (Hill-Jarrett et al., 2015). Unfortunately, however, the pattern of N1 activation was similar in both groups, suggesting that PA patients may preserve adequate orienting function. Retinal eccentricity modulated attentional demands (Beaton and Blakemore, 1981; Asanowicz et al., 2012), and larger eccentricities may make targets difficult to discriminate and require additional attentional resources. Therefore, in the present study, cues and target stimuli may not be presented peripherally enough for allocating resources in orienting. The greater horizontal distance of stimuli presentation and increased differentiation of orienting networks between groups may be beneficial for the further investigation in functional status of orienting networks in PA patients.

Lastly, a significant negative correlation between serum PRL levels and executive functions was found in prolactinoma patients, suggesting the toxic effect of hyperprolactinemia on cognitive functions (Yao et al., 2017; Chen et al., 2022). Meanwhile, the absence of lateralization in the PA group may result from compound factors that mainly contained the compensatory effect of LH and the destructive effect of hyperprolactinemia. The former enhanced the recruitment of attentional resources, and the latter may attenuate the standard lateralized activation pattern and reduce the efficiency of the executive control network.

The present study provided behavioral and electrophysiological evidence of alteration in lateralized attention networks evoked by the LANT paradigm in patients with pituitary adenoma. No ERP differences were found in alerting and orienting network, except for a specialized pattern of contralateral activation in orienting N1 component. More importantly, the executive network P3 exhibited lateralization to RH, and decreased P3 amplitude in the PA group revealed impairment of inhibition control and reduced attentional resource allocation. Moreover, attenuated hemispheric asymmetry of P3 was observed in PA patients, which may be attributed to the mixed effect including the compensatory recruitment of attentional resources in LCP and the destructive effects of hyperprolactinemia. These findings suggested that, in the lateralized condition, the decreased P3 in the RCP and the diminished hemispheric asymmetry under high conflict load, may serve as the potential biomarkers of attentional dysfunction in patients with pituitary adenoma. In addition, the findings above provide further evidence for functional recovery in post-surgery pituitary adenoma patients.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by the Ethics Committee of the General Hospital of Central Theater Command. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

SW proposed and performed the present study, and contributed to collecting and analyzing the experimental data, as well as writing the manuscript. ZF reviewed the manuscript and suggested article revisions. YS contributed to collecting data and writing code. MZ provided clinical guidance. AC and CC contributed to designing the experiment. JS supervised and funded the present study. All authors contributed to the article and approved the submitted version.

The funding of the present study was provided by the National Natural Science Foundation of China (81870863).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.