Functional connectivity changes in the intra- and inter-brain during the construction of the multi-brain network of pigeons

Highlights

• Three pigeons performed social cooperation tasks by using neural feedback training.

• They can achieve cross-brain neural synchronization after training.

• The intra- and inter-brain functional connectivity enhance with the training time.

• This phenomenon is related to the promotion of social cooperation among pigeons.

Abstract

Multi-brain network, also known as social cooperative network, is formed by multiple animal or human brains, whose changes of functional connectivity in the intra- and inter-brain during construction are unclear at present. To investigate the intra- and inter-brain functional connectivity of pigeons while performing a social cooperation task, we designed an inter-brain synchronization task to train three pigeons to synchronize their neural activities using cross-brain neurofeedback. Then the neural signals of three pigeons were simultaneously recorded by using a hyperscanning approach, and inter-brain synchronization was calculated using the phase-locked value (PLV) online. Finally, the intra- and inter-brain functional connectivity of three pigeons were analyzed. We found that during long-term neurofeedback training, with the increase of the inter-brain synchronization of three pigeons, the intra- and inter-brain functional connectivity also enhance significantly. Moreover, we also found that the above phenomenon relies on the external visual cue. These results suggest that the promotion of social cooperation is the result of the modulation between the intra- and inter-brain, which may be an underlying neural mechanism of the communication and cooperation among individuals in social networks.

Introduction

In nature, humans and other animals do not exist independently but are embedded in a social network for daily communication and cooperation. Cooperation, in particular, can be defined as a social interaction between two or more agents that can improve work efficiency. In fact, cooperation is considered as an interpersonal interaction that supports joined actions and that is able to affect the immediate and future behavior of the other individuals involved in the exchanges. Recent social neuroscience approaches point out that cooperation can only be better understood by considering interactive participants as a unique system (Hasson et al., 2012). However, one limitation of previous studies on the neurological basis of social interaction is that they only focus on the intra-brain of individuals, but neglect the interaction between individuals (Spiegelhalter et al., 2014).

Therefore, new research needs to shift towards multi-brain neuroscience (Schilbach et al., 2013). In recent years, hyperscanning technique applied to simultaneously record the neural activities from different multiple brains has been widely used to explore the neural mechanisms of social interaction (Takahiko et al., 2015, Xu et al., 2012). In fact, using this technology, the researchers found that multiple participants in the various social interaction tasks show the phenomenon of cross-brain neural synchronization, such as body movement coordination and imitation (Holper et al., 2012, Yun et al., 2012), audience-engagement during listening to political speeches (Schmälzle et al., 2015), interactive decision-making (Hu et al., 2018), cooperation keypress (Xu et al., 2012, Kawasakia et al., 2018) and face-to-face communication (Jing et al., 2012, Ahn et al., 2017). These interesting results suggest that the phenomenon of cross-brain neural activity synchronization is widespread in daily life, which may play a key role in the social activities of humans and animals.

Furthermore, when cooperating, the adoption of common strategies is crucial and the individuals self-perceived may be influenced by others from the system to present a better synergistic effect. Previous experiments already investigated the effects of external feedback by assessing the behavioral performance for cooperative or competitive tasks (Balconi and Vanutelli, 2016). In fact, in some cases, the positive cooperative effect is broken due to the perception of ineffective outcomes of our own actions. So effective information feedback can be considered a powerful cue that can reciprocally reinforce or weaken the behavior toward a common goal, and a relevant tool to train the brain to work synergically or more "individually" to compensate the cognitive gap. In fact, neurofeedback can also be regarded as "brain training", which is an effective method to investigate brain activity. During the brain training process, the neural signals of the subjects are observed and extracted in real-time and fed back to the subjects in the form of vision and hearing, who can selectively enhance or inhibit certain components by training to achieve the purpose of regulating brain function (Mcwhinney et al., 2017). Evidence has shown that in the past decade, subjects can improve their cognitive or behavioral abilities by using long-term neurofeedback technology to regulate brain activity (Kazuhisa et al., 2011, Weiskopf, 2012, Kopel et al., 2017, Prins et al., 2017). Moreover, Duan et al. (2013) extended neurofeedback from single-person context to the multi-person situation. This work allows multiple participants simultaneously self-regulate their own neural activities in a social interacting situation.

Up to now, cross-brain neurofeedback has been used to construct various forms of Brainet in mammals such as humans, monkeys, rats and so on. Nicolelis et al. first proposed the concept of Brainet (Nicolelis, 2011) in 2011, which referred to a network composed of multiple interconnected animal brains, and experimentally proved that the multi-brain network can self-adjust to achieve common moving targets (Ramakrishnan et al., 2015), real-time cooperation and exchange of information (Paisvieira et al., 2013, Pais-Vieira et al., 2015). In addition, Duan et al. (2013) used cross-brain neurofeedback to train two subjects to complete the virtual tug-of-war task. The results showed that the inter-brain synchronization was significantly improved by neurofeedback training. Not only that, Ramakrishnan et al. (2015) trained three monkeys to grasp the ball by adjusting their neural signals to control the movement of the manipulator's arm. In the same year, they continued to improve the experimental paradigm, interconnecting the brains of four rats to build a Brainet to obtain water reward by synchronizing their neural signals (Pais-Vieira et al., 2015). Moreover, Jiang et al. (2019) successfully constructed a BrainNet composed of multiple human brains using a brain-to-brain interface (BBI) cooperative system for the first time. The above results show that the brains of primates can be integrated into a Brainet using neurofeedback to self-adjust to achieve common goals of the movement, and also indicate that the construction of Brainet requires long-term induction to synchronize the neural activities of multiple brains.

However, in the process of constructing the multi-brain network, how the functional connections of the intra- and inter-brain change, and whether these changes are related to the promotion of social activities are unclear. Here, we supposed that the intra- and inter-brain functional connectivity may be affected by external social feedback, which may be associated with improvements in group social cognition in a joint task. In addition, this increased connectivity might reveal a positive attitude toward effective cooperation by the inter-agents. Pigeon is a typical gregarious animal (Dell’Ariccia et al., 2008, Nagy et al., 2010), which is an ideal model animal to explore the relationship between social cooperative behavior and inter-brain neural activity. Recent researches suggest that avian nidopallium caudolateral (NCL) resembles the mammalian prefrontal cortex (PFC) in anatomical connectivity, neurochemical organization, receptor architecture, and functional characteristics (Kröner and Güntürkün, 1999, Bast et al., 2002, Kalt et al., 1999, Kirsch et al., 2009), which has been previously demonstrated to be sensitive to and relevant in social and cooperative exchanges (Liu et al., 2016). Here, we expected that the inter-brain coupling modulation in response to effective cooperation should involve a specific contribution by the NCL of the pigeon.

Therefore, the present study aims at moving towards a multi-brain hyperscanning perspective and attempting to construct a synchronized multi-brain network composed of the pigeons. In details, we train the pigeons to achieve inter-brain synchronization and integrate multiple brains into one brain to perform a joint attention task, to investigate the intra- and inter-brain functional connectivity and the inter-brain synchronicity by LFP measure recorded in the joint attention task. The results will help to understand the neural mechanisms of animal populations’ social behavior and to provide further support for social learning in the complex environment in nature.