Language is tightly interwoven with interpersonal cognition (Tylén et al., 2010; Seyfarth and Cheney, 2014; Corballis, 2017). Using approach and avoidance verbs, language enables us to represent and communicate our own attitudes, as well as those of others, in social contexts. This helps regulate our social interactions by expressing preferences or aversions (Marrero et al., 2023a,b). In social settings, we express our preferences using attitudinal expressions, often these preferences are indexed by verbs that denote approach or avoidance (Marrero et al., 2020, 2023a,b). Avoidance verbs suggest a negative stance toward stimuli. Early in infancy, we use both bodily and linguistic signs, like head shaking or verb denial, to show our dislike for certain stimuli (Clark, 1978; Austin et al., 2014; Beaupoil-Hourdel et al., 2016). Over time, verbal avoidance action may become represented as “against a stimulus approach”; given that “against” is perceived as negative (Clark, 1978; Xiang et al., 2016), avoidance verbs inherently would be represented as implicitly negative, like “exclude” being understood “not include” (Marrero et al., 2020; see also Dudschig et al., 2023 and Montalti et al., 2024).

Negation is an important linguistic operator (Horn, 1989), which can also serve social communication. Using questionnaires, the role of negation in reversing the meaning of attitudinal sentences from approach to avoidance, and vice versa, has been demonstrated off-line (Marrero et al., 2020). Given its cognitive and social relevance, it would be of interest to examine the brain’s basis for approach and avoidance and the interplay with negation in verbal understanding for interpersonal cognition.

The right Inferior Frontal Cortex (rIFG) has been shown to be involved in inhibitory processes, both motor and cognitive (Aron et al., 2014). Since linguistic negation induces inhibition of the negated state of affairs, such as “an open door” in the utterance “the door is not open” (Kaup, 2006; Liuzza et al., 2011; Bartoli et al., 2013; Foroni and Semin, 2013; García-Marco et al., 2019; Kaup and Dudschig, 2020; Marrero et al., 2020; Papitto et al., 2021; Montalti et al., 2023, 2024), several studies have demonstrated the involvement of right Inferior Frontal Gyrus in negation processing (de Vega et al., 2016; Papeo et al., 2016; Beltrán et al., 2018, 2019, 2021; Liu et al., 2020; Vitale et al., 2022). For example, inhibitory stimulation of rIFG by rTMS (repetitive transcranial magnetic stimulation) suppressed the inhibitory effect of negation on the action representation, equalizing motor excitability between negative and affirmative action sentences. This shows negation has a “disembodiment” effect that is reflected in reduced motor excitability during negated action comprehension (Liuzza et al., 2011; Bartoli et al., 2013; Papitto et al., 2021; Vitale et al., 2022).

Research has explored whether brain stimulation in the right temporal area, specifically in the medial aspects of the Superior Temporal Sulcus (STS), affects the processing of negation in attitudinal approach/avoidance sentences (Nuez et al., 2022). Participants read sentences such as: “Anne included (did not include) meat in her diet,” “Anne excluded (did not exclude) meat in her diet.” Then, 1,500 ms after the sentence display, they read a verb either mentioned in the sentence or its alternative (e.g., “include” or “exclude”). Results revealed that anodal stimulation increased availability of approach verbs (like “include”), indexed by shortened reading times. Additionally, this stimulation also made negated verbs more available. These findings suggest this right temporal area specializes in factual and approach information processing.

The so-called two-step model stands as the most widely accepted model to explain the processing of negation in language (Kaup, 2006; Kaup et al., 2006; Dudschig and Kaup, 2018; Kaup and Dudschig, 2020; Beltrán et al., 2021). According to this model (Kaup, 2006), comprehension of negative sentences happens in two sequential steps. First, negation focuses attention on what is negated (e.g., an “open door”). Following this, in a second step, this situation is dismissed, leading to an inhibited representation, evidenced by reduced accessibility of either an associated word or image in a recognition task. In this step, alternate scenarios (e.g., a closed door in the example above) might emerge. This is especially the case for binary categories (like even/odd) or contradictory terms (like open/closed) (Kaup et al., 2006). Within the two-step model framework, the effect of tDCS stimulation on medial rSTS in negation processing appears to occur in the first step, enhancing the effect of negation on availability of the negated element (Nuez et al., 2022; Vitale et al., 2022).

A preliminary comparison of verb reading response times pre-tDCS stimulation revealed no significant disparities between the anodal and sham groups, with p > 0.10. A linear mixed model fitted by restricted maximum likelihood adjusted for the response time distribution was applied to the experimental data. The t-tests used the Satterthwaite method. We utilized the Holm method, a stepwise procedure designed to control the family-wise error rate in a conservative manner.

Although the four-way interaction direction x polarity x verb x tDCS did not reach a significance level (β = −0.23, t = −1.5), further examination of the predicted effects is pertinent to explore the interactions that arise separately for each type of tDCS stimulation (anodal vs. sham).

A summary of Linear mixed-effects model for the anodal condition is shown in Table 3.

Target-verb reading times across different conditions is shown in Table 4. We found a main effect of Direction (β = 0.083, t = 3.2); target-verbs were read faster in avoidance sentences (M = 769) than in approach sentences (M = 805). Additionally, we found a main effect of Verb (β = 0.061, t = 2.4); mentioned verbs showed shorter reading times (M = 766) than alternative verbs (M = 809). The polarity x verb interaction was significant (β = 0.14, t = 2.8). Post-hoc comparisons showed that mentioned verbs presented longer reading times in negative (M = 785) than in affirmative sentences (M = 747, β = −0.0912, SE = 0.0367, z = 2.482 = 2.4, p = 0.01). Conversely, in affirmative sentences, alternative verbs showed longer reading times (M = 825) than mentioned verbs (M = 747) (β = −0.1334, SE = 0.0367, z = 3.63, p = 0.003).

This interaction was qualified by a significant three-way direction x polarity x verb interaction (β = −0.3, t = −2.9). Pairwise comparisons showed that alternative verbs were read more slowly in approach negative sentences than in avoidance negative sentences (β = 0.2411, SE = 0.052, z = 4.64, p < 0.001). Alternative verbs showed longer reading times in avoidance affirmative sentences than in avoidance negative sentences (β = 0.1731, SE = 0.0521, z = 3.233, p < 0.001). The reading time of mentioned verbs was significantly shorter in avoidance affirmative sentences than in avoidance negative sentences (β = 0.1208, SE = 0.0521, z = −2.317, p = 0.02). Alternative verbs presented longer reading times than mentioned verbs in avoidance affirmative sentences (β = 0.1696, SE = 0.0521, z = 3.25, p = 0.001); and they presented longer reading times than the mentioned verbs in approach negative sentences (β = 0.1021, SE = 0.0518, z = 1.971, p = 0.0487). By contrast, alternative verbs presented shorter reading times than mentioned verbs in avoidance negative sentences (β = −0.1243, SE = 0.0521, z = 2.386, p = 0.0170). Neither the main effect of Polarity (β = −0.019, t = −0.73) nor the direction x verb interaction (β = 0.077, t = 1.5) were significant.

Figure 3 illustrates the effects of negation on reading times of mentioned verbs (a) and alternative verbs (b), which supports predictions from H1 and H2.

In the sham condition (Table 5), the triple interaction direction x polarity x verb was not significant (β = −0.069, t = −0.62). Similarly, none of the main effects of Direction (β = 0.0041, t = 0.1), Polarity (β = −0.044, t = −1.6), and Verb (β = 0.037, t = 1.3) were significant. In addition, the direction x polarity (β = 0.032, t = 0.58), direction x verb (β = 0.073, t = 1.), and polarity x verb (β = 0.014, t = 0.25) interactions were not significant.

Overall, the results support our hypotheses. In the anodal condition, the polarity x verb interaction supports H1, indicating longer reading times for mentioned verbs in negative than affirmative sentences. In the sham condition, there were no significant reading time differences for the mentioned verbs based on polarity. This suggests that anodal stimulation of the rIFG amplifies the inhibitory effect of negation for mentioned verbs. In addition, the three-way interaction direction x polarity x verb supports H2, alternative verbs in anodal condition resulted in shorter reading times in negative avoidance sentences compared to negative approach sentences.

These results contribute to our understanding of the brain’s basis of interaction of negation and approach-avoidance dynamics in linguistic and social communication. The different patterns of results between the anodal and sham conditions underline the rIFG pivotal role in the inhibitory effect over linguistic negation. The negation effect is different depending on whether the verbs are mentioned or alternative, and whether the sentences are approach or avoidance. For mentioned verbs, an amplified negation effect manifests through pronounced inhibition, indexed by less accessibility and longer verb-reading times, in negated sentences compared to affirmative sentences, either of approach or avoidance. This confirms our prediction for the second step of negation processing (Kaup, 2006; Kaup et al., 2006; Dudschig and Kaup, 2018; Beltrán et al., 2019; Kaup and Dudschig, 2020; Beltrán et al., 2021). For alternative verbs, the effect of negation depends on the direction of the verb, whether the sentence is of approach or avoidance. For avoidance verbs, anodal stimulation also enhances the effect of explicit negation, but in this case the effect is exerted on the implicit negation, e.g., NOT excluded = NOT→NOT (included), thereby releasing the representation of the alternative approach verb (“included”). This result suggests that the amplified inhibitory effect of negation, driven by right IFG brain stimulation, can be exerted over an implicit negation, enhancing the accessibility of the alternative scenario. However, its availability is not necessarily associated with inferring the implications of negation, as predicted by the two-step model.

Previous research has found that anodal tDCS stimulation of the right temporal lobe (around medial aspects of the STS) enhances the effect of negation through greater accessibility of what is negated (Nuez et al., 2022). If we compare the anodal stimulation in the right IFG with that of the right superior temporal sulcus, it suggests that negation processing could be performed in a sequential manner involving different brain areas, which could constitute a specialized brain network following the two-step model. The right temporal area would specialize in processing approach and factual nuances in language, and negation would be processed within the first step where attentional focus is on what is negated. In contrast, the right IFG would specialize in processing the negation function of inhibition, in the second step, which would give rise to a new and relevant effect in avoidance utterances where inhibition would be exerted in an implicit negation.

This research explores the interplay between approach-avoidance language and negation processing and understanding the brain’s basis of these processes. Approach and avoidance attitudinal verbs play a pivotal role in the interwoven of language with interpersonal cognition (Tylén et al., 2010; Seyfarth and Cheney, 2014; Corballis, 2017), facilitating the communication of our own and others’ attitudes in social settings, and thereby regulating interpersonal interactions. Negation has long been recognized as an essential linguistic operator in social communication (Horn, 1989). The results of this study provide scientifically relevant knowledge in this regard.

Indeed, our results support the role of the rIFG in inhibition and negation processing from previous studies (de Vega et al., 2016; Papeo et al., 2016; Beltrán et al., 2018, 2019, 2021; Liu et al., 2020; Vitale et al., 2022). Moreover, we examine, for the first time, the communication of attitudes through language and its interaction with negation during comprehension. In particular, this study has investigated social avoidance in language where verbs are negatively implicit as well as highlighting the relevance of rIFG in avoidance processing.

Stimulation with tDCS has certain limitations. For rIFG stimulation, we adhered to the 10/20 EEG system for electrode placement. However, factors such as inter-subject anatomical variability and the potential lack of focality in the applied stimulation could have influenced the results. Therefore, future research endeavors employing other techniques such as fMRI or TMS that could provide more direct and precise evidence their inhibitory role in negation processing. Also important, the four-way significant interaction did not result significant which weakens the contrast among anodal and sham group to support our hypothesis. A plausible explanation is that in the sham group we had not enough power. Future research is necessary for replication of the study by increasing samples, in particular of the sham group.

Our participant cohort predominantly comprised young, female university scholars. However, approach and avoidance brain encoding could be affected by developmental changes or be modulated by gender (Pintzinger et al., 2016; Torrence and Troup, 2018). Thus, future studies should also include a more heterogeneous demographic.

This study examines the effect of anodal tDCS on the rIFG in processing negation and social avoidance in language. Avoidance attitudinal verbs like “exclude” are implicitly negatively represented as “no-approach.” We found that anodal tDCS stimulation enhanced inhibitory effect of negation in two different ways. Firstly, longer reading times were shown in mentioned verbs in negated sentences compared to affirmative sentences. This result supports an inhibitory effect of negation on the verb mentioned, consistent with the second step of negation processing (Kaup, 2006; Kaup et al., 2006; Dudschig and Kaup, 2018; Kaup and Dudschig, 2020; Beltrán et al., 2021). Secondly, we found shorter reading times of alternative verbs in negative avoidance compared to negative approach sentences in anodal condition. This suggests that the enhanced inhibition from explicit negation was exerted over the implicit negation in avoidance sentences. To our knowledge, this provides novel evidence that an inhibitory effect of negation can be exerted on implicit negations, beyond its well-known effect on inhibition of narrated events. Moreover, this negation effect on a double negation highlights the role of rIFG in processing social avoidance meaning in language. Finally, we further theorize whether the rIFG might be part a brain network for processing the meaning of negation (Ghio et al., 2018) joined to right temporal area (specifically, medial aspects of the STS) that could specialize in the first step. Further research is necessary for examining this hypothesis.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by the ethical committee of the University of La Laguna approved the study: (CEIBA 2017-0272). 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.

EG-M: Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing. AN: Project administration, Supervision, Writing – original draft, Writing – review & editing. IP: Data curation, Formal analysis, Methodology, Writing – original draft, Writing – review & editing. YR: Data curation, Writing – original draft, Writing – review & editing. YF: Writing – original draft, Writing – review & editing, Data curation, Formal analysis. HM: Project administration, Supervision, Writing – original draft, Writing – review & editing.