INTRODUCTION
Technology innovations may be of various forms (i.e., disruptive, incremental, radical, or architectural). Regardless of the types, technologies support our daily tasks. Among the many recent trends in innovations, the “Like” button (feature) is considered a type of disruptive innovation for businesses (Sullivan, 2016).
Numerous social media platforms (e.g., Facebook, YouTube, Twitter, LinkedIn, Instagram, TikTok) introduced the “Like” feature (in one form or another) to solve problems and save time, but it ended up creating more challenges and unintended consequences. Numerous empirical and theoretical studies focused on the perceived outcomes of the “Like” feature (e.g., intention, service quality) (e.g., John, Emrich, Gupta, & Norton, 2017; Schondienst, Kulzer, & Gunther, 2012), on its social aspects (Eranti & Lonkila, 2015), and the motives behind its specific or general use (e.g., Ozanne, Navas, & Mattila, & Van Hoof, 2017; Kim, Sohn, & Choi, 2011).
Nevertheless, despite the growing body of knowledge on social media use (generally) and the “Like” feature (specifically), there is a dearth of research focusing on the underlying behavioral mechanisms of the “Like” feature, especially since academic publications in this research area are relatively scarce. Research studies conducted by health experts (e.g., Royal Society for Public Health, 2018) have shown that the “Like” button is the most toxic feature on social media by triggering undesirable reactions, emotions, memories, stress, and pressure (Moffat, 2019). Such behavioral experiences/reactions are challenging to be measured by traditional/commonly used methodologies (e.g., interviews, surveys), which led to inaccurate behavioral mappings and unreliable findings. One approach to effectively measure behavioral (neurophysiological) reactions to the “Like” button is through the neuroIS perspective, which this research adopted. NeuroIS refers to applying cognitive neuroscience methods and tools in IS research (Dimoka, Pavlou, & Davis, 2007).
The “Like” feature, a simple yet innovative method of gathering data of individuals’ interests and activities, has shown to be beneficial for businesses but damaging to users’ well-being (Moffat, 2019). Yet, no study in the literature investigated users' behavioral responses (from a neurophysiological perspective) towards the ‘’Like’’ feature. Thus, this research attempted to address “How do users neurophysiologically behave or respond towards the ‘’Like’’ feature?”
This research delivers three modest contributions. First, neuroIS research is relatively new and underdeveloped (Riedl, Fischer, Leger, & Davis, 2020). Therefore, through the theoretical lens of cybernetics, this neuroIS research is the first in the literature to investigate how users neurophysiologically behave or respond towards the “Like” feature. This research experimented with two different neurophysiological tools (i.e., electrocardiogram (EKG/ECG) and electroencephalography (EEG)). Second, this research delivered a simplified yet comprehensive understanding of users’ neurophysiological responses by challenging the complexities (e.g., costly and complicated experimental designs) of previous neuroIS experimental studies; thus, capturing a broader range of audiences. Besides, such an attempt supports higher potentials of replicability and reproducibility to develop neuroIS research. Third, to date and up to the authors’ knowledge, two neuroIS studies investigated users responses towards the ‘’Like’’ feature but with the use of fMRI rather than EEG or EKG/ECG (i.e., Sherman, Hernandez, Greenfield, & Dapretto, 2018; Sherman, Payton, Hernandez, Greenfield, & Dapretto, 2016). However, fMRI focuses on spatial (which area of the brain is active) rather than temporal resolution (when activation occurs). The temporal resolution of fMRI has shown to be inaccurate due to the rapid circulation of blood (Fomby & Cherlin, 2011). Besides, fMRI is more fit for studies related to memory and cognition (Pandarinathan et al., 2018). Accordingly, this research provided a novel and better understanding of users’ responses towards the ‘’Like’’ feature from the temporal perspective. Subsequently, a complete road map of both resolutions (spatial and temporal) will be available for future studies to build on.
Furthermore, the “Like” feature tracks and monitors users through pre-determined algorithms (Reflectiz, 2019; Eranti & Lonkila, 2015). As a result, throughout the last decade, the “Like” feature drew privacy scrutiny (i.e., sharing user information/data to third parties). This research further raised concerns over the underlying AI algorithms that partially function within “recommendation engines’’.
The rest of the research is structured as follows: the following section reviews the theoretical backgrounds on the concept of user behavior, social media in the information systems (IS) discipline, the “Like” feature, neuroscientific methods, and cybernetic theory. Then, the authors discuss the methodology section that involves a detailed experimental protocol. The authors then elaborate on the findings and conclude with limitations and future research directions.