Language occurs naturally in conversations. However, the study of the neural underpinnings of language has mainly taken place in single individuals using controlled language material. The interactive elements of a conversation (e.g., turn-taking) are often not part of neurolinguistic setups. The prime reason is the difficulty to combine open unrestricted conversations with the requirements of neuroimaging. It is necessary to find a trade-off between the naturalness of a conversation and the restrictions imposed by neuroscientific methods to allow for ecologically more valid studies. Here, we make an attempt to study the effects of a conversational element, namely turn-taking, on linguistic neural correlates, specifically the N400 effect. We focus on the physiological aspect of turn-taking, the speaker-switch, and its effect on the detectability of the N400 effect. The N400 event-related potential reflects expectation violations in a semantic context; the N400 effect describes the difference of the N400 amplitude between semantically expected and unexpected items. Sentences with semantically congruent and incongruent final words were presented in two turn-taking modes: (1) reading aloud first part of the sentence and listening to speaker-switch for the final word, and (2) listening to first part of the sentence and speaker-switch for the final word. A significant N400 effect was found for both turn-taking modes, which was not influenced by the mode itself. However, the mode significantly affected the P200, which was increased for the reading aloud mode compared to the listening mode. Our results show that an N400 effect can be detected during a speaker-switch. Speech articulation (reading aloud) before the analyzed sentence fragment did also not impede the N400 effect detection for the final word. The speaker-switch, however, seems to influence earlier components of the electroencephalogram, related to processing of salient stimuli. We conclude that the N400 can effectively be used to study neural correlates of language in conversational approaches including speaker-switches.

After the experiment, the EEG cap was removed and participants were asked to fill in an evaluation questionnaire. Participants were for example asked to rate how natural the interaction seemed, how pleasant it was, and how pleasant the interaction partner was on a five-point scale from 'not at all' to 'very' (see Supplementary-Evaluation Results). A complete experimental session lasted around 3 to 3.5 hours. EEG recording. Brain electrical activity was measured with a 96-channel EEG system (BrainProducts, Gilching, Germany). Ag/AgCl electrodes were placed equidistantly with a nose tip reference and centro-frontal ground (Easycap, Herrsching, Germany). Impedances were kept below 20 kΩ. Data were digitized with a sampling rate of 500 Hz. EEG stream, marker stream, and audio streams were recorded synchronously using the Lab Recorder from Lab Streaming Layer 59 .

We introduce here the word-by-word paradigm, a dynamic setting, in which two people take turns in producing a single sentence. This task requires a high degree of coordination between the partners and the simplicity of the task allows us to study with sufficient experimental control behavioral and neural processes that underlie this controlled interaction. For this study, 13 pairs of individuals engaged in a scripted word-by-word interaction, while we recorded the neural activity of both participants simultaneously using wireless EEG. To study expectation building, different semantic contexts were primed for each participant. Semantically unexpected continuations were introduced in 25% of all sentences. In line with the hypothesis, we observed amplitude differences for the P200-N400-P600 ERPs for unexpected compared to expected words. Moreover, we could successfully assess speech and reaction times. Our results show that it is possible to measure ERPs and RTs to semantically unexpected words in a dyadic interactive scenario.

Social interaction plays an important role in many contexts of human reasoning and problem solving, and groups are often found to outperform individuals. We suggest that this benefit is associated with the dialogical sharing and integration of diverse perspectives and strategies. Here, we investigated whether diversity in prior experience affects groups’ problem representations and performance. In a game-like experiment, participants categorized aliens based on combinations of their features. Whenever a specific feature combination was learned, the rule changed and a new feature combination had to be learned. However, unbeknown to participants, rule changes were governed by an abstract meta-rule and awareness of this provided an advantage when rules changed. We compared categorization performance between individuals, groups composed of members trained on the same rule, and groups composed of members trained on different rules before entering the collaborative test phase. Following preregistered predictions, groups with diverse task experience outperformed groups with similar task experience, which in turn outperformed individuals. These findings were unaffected diversity in personality (Big Five) and motivational factors, suggesting that diversity in experience plays the key role. We conclude that cognitive diversity impact problem solving by stimulating processes of abstraction and flexibility at the level of the group.

Cognitive flexibility is the ability to switch between different concepts or to adapt goal-directed behavior in a changing environment. Although, cognitive research on this ability has long been focused on the individual mind, it is becoming increasingly clear that cognitive flexibility plays a central role in our social life. This is particularly evident in turn-taking in verbal conversation, where cognitive flexibility of the individual becomes part of social flexibility in the dyadic interaction. In this work, we introduce a model that reveals different parameters that explain how people flexibly handle unexpected events in verbal conversation. In order to study hypotheses derived from the model, we use a novel experimental approach in which thirty pairs of participants engaged in a word-byword interaction by taking turns in generating sentences word by word. Similar to well established individual cognitive tasks, participants needed to adapt their behavior in order to respond to their co-actor's last utterance. With our experimental approach we could manipulate the interaction between participants: Either both participants had to construct a sentence with a common target word (congruent condition) or with distinct target words (incongruent condition). We further studied the relation between the interactive Word-byWord task measures and classical individual-centered, cognitive tasks, namely the Number-Letter task, the Stop-Signal task, and the GoNogo task. In the Word-byWord task, we found that participants had faster response times in congruent compared to incongruent trials, which replicates the primary findings of standard cognitive tasks measuring cognitive flexibility. Further, we found a significant correlation between the performance in the Word-byWord task and the Stop-Signal task indicating that participants with a high cognitive flexibility in the Word-byWord task also showed high inhibition control.