Does the Language You Learned Change How Your Brain Pays Attention?

When a word contradicts our expectations, does the brain need to reconstruct meaning? What can syntax and humor reveal about this process?

Language · Predictive Processing · EEG/ERP · P600 · Humor · Body-Territory · Metanoia

Imagine hearing someone say:

“The boy met...”

Before the sentence is complete, you already anticipate possibilities. Another person? An animal? Someone familiar? Your brain begins constructing possible meanings without knowing what comes next.

Now consider a joke:

“— Doctor, everyone ignores me.
— Next!”

The humor emerges when we realize that the doctor's response demonstrates exactly what the patient was complaining about.

In both cases, something fundamental happens: the brain anticipates an interpretation and may need to revise it when new information arrives.

But could understanding unusual sentence structures and comprehending jokes involve shared neurophysiological processes?

This question emerges from research by Medine Dudu, Yağmur Köyel Yıldırım, and Talat Bulut, published in October 2026 in Frontiers in Human Neuroscience: Processing canonical and non-canonical word orders in Turkish: behavioral and ERP evidence from adults and children.

1. Comprehension begins before a sentence ends

When we hear words, our auditory system processes acoustic information, recognizes phonetic regularities, and identifies learned linguistic representations.

However, comprehension does not mean passively assembling phonemes into words and then words into sentences.

Context participates from the beginning.

Earlier information influences what we expect to hear next. An unexpected word can require adjustments to an interpretation already under construction.

During reading, written symbols provide another sensory input, but anticipation and integration remain important.

BrainLatam explored this relational dimension in Before Words — The Other Already Participates in the Architecture of Language.

The experimental question is: what happens when sentence structure requires us to revise our expectations about who performs an action?

2. Subject and object: word order changes across languages

In English, a common word order is Subject–Verb–Object (SVO):

“The boy hugged the girl.”

In Turkish, the canonical word order is Subject–Object–Verb (SOV).

However, Turkish also permits Object–Subject–Verb (OSV) constructions.

Morphological markers, including accusative case, help identify who performs and who receives an action, even when participants change positions.

This makes Turkish particularly interesting for investigating syntactic expectations.

Subject and object remain identifiable, but the system must integrate grammatical information presented in sequences with different frequencies.

Non-canonical structures can also serve legitimate discourse functions, such as highlighting previously known information.

Therefore, an unusual word order is not necessarily incorrect.

3. How did Dudu and colleagues investigate this difference?

The study included 67 adults aged 18–44 and 37 children approximately aged 4–10, all native Turkish speakers.

Participants listened to sentences and then selected which of two pictures correctly represented the described event.

In one picture, for example, a police officer was painting a doctor. In the other, the doctor was painting the police officer.

The sentences used the same characters and verbs but changed word order and the grammatical markers identifying their roles.

The researchers constructed 48 experimental sentence sets containing SOV and OSV versions.

During the task, they recorded brain electrical activity using Brain Products actiCHamp and actiCAP equipment, with 27 scalp EEG electrodes and a sampling rate of 1,000 Hz.

Their analysis focused on event-related potentials (ERPs), time-locked to the onset of the final verb.

This approach combined behavioral accuracy with precise temporal measurements of brain activity.

4. What did the results reveal?

OSV sentences produced lower comprehension accuracy in both groups.

Among adults, the difference relative to SOV sentences was approximately 2.77 percentage points.

Among children, it reached 8.91 percentage points.

The behavioral disadvantage associated with non-canonical word order was significantly greater among children.

This shows that recognizing grammatical markers does not guarantee equally efficient integration across different sentence structures.

In adults, OSV sentences also produced a larger late posterior positivity than SOV sentences, with significant effects across intervals between approximately 600 and 1,200 milliseconds after final-verb onset.

Among children, a corresponding positivity was not statistically reliable after correction. However, direct comparisons between groups did not establish a significant electrophysiological difference.

The researchers also found that stronger grammatical abilities were associated with greater comprehension accuracy among children.

These findings do not demonstrate differences in consciousness between children and adults. They reveal different processing demands associated with linguistic structure.

5. A temporal window that also appears in jokes

Here we encounter a particularly interesting research opportunity.

In Dudu and colleagues' investigation, late posterior positivity appears when adults integrate a less habitual syntactic structure.

But studies of verbal humor have also identified late electrophysiological responses during the reinterpretation of unexpected meanings.

In Getting it: Human event-related brain response to jokes in good and poor comprehenders, Seana Coulson and Marta Kutas investigated how readers understand jokes, identifying responses associated with surprise and shifts in interpretive framing.

More recently, Fan Peng, Rong Zhou, and Musi Xie published L2 exposure intensity modulates the time course of verbal humor comprehension: An ERP study in 2026.

In that experiment, 42 students learning English as a foreign language read jokes and coherent endings.

The researchers observed larger N400 responses to humorous endings and a P600 response approximately between 500 and 900 milliseconds among participants with greater exposure to English.

These findings suggest that linguistic experience influences later reinterpretation processes.

The temporal similarity with syntactic processing is intriguing, but it does not demonstrate that identical neural circuits perform both operations.

6. Could predictive revision be the shared mechanism?

We can organize this hypothesis into four stages.

First, previous experiences guide expectations about what is likely to happen.

Second, new information contradicts or complicates the initial interpretation.

Third, the system must integrate additional cues or reinterpret relationships.

Finally, an updated representation allows comprehension to continue.

In OSV sentences, this process involves grammatical roles.

In jokes, the punchline may require reinterpretation of the entire situation.

The N400 is commonly associated with aspects of semantic processing, including contextual expectations and integration. The P600 and other late positive responses may accompany additional reanalysis and integration operations.

However, these ERP components are not exclusive indicators of specific cognitive stages.

Similar late electrical responses do not necessarily reflect the same neural mechanism.

The scientific challenge is to determine which operations are shared and which depend specifically on syntax or humor.

7. The language we learn contributes to our expectations

If linguistic experience shapes expectations, we can investigate how different languages influence the processing of events.

Turkish allows comparisons between SOV and OSV structures.

English and Portuguese frequently organize sentences using SVO.

Indigenous Latin American languages, including varieties of Quechua, offer further opportunities to investigate syntax, morphology, and discourse context.

These differences do not justify claiming that one language produces a superior or inferior form of consciousness.

We must also distinguish grammatical subjects and objects from the philosophical subject who experiences and the object being experienced.

Connecting these dimensions generates questions, not automatic equivalences.

Cross-cultural investigations could examine how speakers anticipate agents, actions, and consequences using EEG, eye tracking, and comprehension measures.

Such research should respect linguistic differences and involve researchers from the communities concerned.

8. From predictive revision to Metanoia of the Metastable Brain

At BrainLatam, we propose distinguishing three levels.

Predictive revision: new information requires an interpretation to change during a particular encounter.

Learning: previous experiences influence performance in later encounters.

Metanoia of the Metastable Brain: a hypothesis concerning more persistent reorganization of an organism's repertoires and possibilities for transitioning between states.

Metastability provides concepts for investigating how neural networks alternate between patterns of coordination without remaining permanently fixed in a single state.

This does not mean that Dudu and colleagues measured metastability or demonstrated permanent transformation.

The proposal is to investigate when repeated revisions of expectations contribute to lasting changes.

Within the Body-Territory perspective, these changes should also be considered in relation to the organism's bodily, social, and cultural experiences.

9. A new experimental question for neuroscience

Imagine an experiment involving canonical sentences, non-canonical sentences, understandable jokes, and unexpected endings without a coherent resolution.

EEG could compare N400 responses and late positivities. Behavioral measures could assess comprehension, surprise, and reinterpretation demands.

The same participants could encounter similar stimuli again after different learning experiences.

The question would be: does revising an expectation change only immediate comprehension, or does it also influence how the next encounter is processed?

This investigation would bring syntax, humor, attention, and learning into dialogue without reducing different phenomena to a single brain response.

Dudu, Köyel Yıldırım, and Bulut's original contribution remains central: linguistic experience and grammatical cues participate in how we understand relationships between individuals and events.

BrainLatam extends this discovery through another possibility:

If understanding requires revising predictions, could learning involve changing the very system that generates those predictions?

In Blog 06, we will move from words to movement: when the body plays, how does attention contribute to coordinating our next actions?

Meta description: Explore how a 2026 EEG study connects word order, predictive processing, P600 responses, and humor comprehension to the brain's ability to revise meaning.

Editorial note: The findings on Turkish sentence processing and verbal humor come from separate studies. Their potential shared predictive-revision mechanisms, as well as the Metanoia of the Metastable Brain and Body-Territory connections, are proposed research questions rather than established findings.

BRAINLATAM | W47/2026 | BLOG 05/12 — ENGLISH VERSION