Why psychologists should care about blockchain data

Blockchains are digital ledgers. Ledgers have been used to record information and transactions for at least 4,000 years. Some of the oldest surviving written records are ledgers carved into clay or written on papyrus. They are typically used to record information like wages, inventories of goods, and records of trade and taxes. Blockchains are digital, distributed versions of a ledger, where every set of new entries into the ledger (clustered into blocks of data) are connected to all previous data in an unbroken chain.

The best way to think of blockchains are as a global network of computers used to maintain a ledger. If you drop your laptop in a pond, the spreadsheet (ledger) saved on your computer is unlikely to survive the ordeal. But blockchain ledgers are continuously maintained by many different computers (nodes) all around the world. This worldwide network of computers are constantly checking that the ledger is accurate and adding new information (as blocks of data) to the ledger. 

I'm assuming most of you reading this are psychologists. So why should you care? Because blockchains create some of the largest databases of open data in existence. They are real-time records of behaviour within a digital network. 

How blockchains work

Blockchains are most commonly associated with cryptocurrency, but there is a key difference. 

Think of blockchain as the operating system, like Android or iOS on smartphones. It's a set of computing rules that anyone can use to create new applications. Cryptocurrencies are like an app that runs on the operating system (like a banking or payment app). The blockchain is the underlying technology that provides the networks that can run various apps. An Android or iOS operating system can support apps like banking, social media, messaging apps and mobile games. A blockchain can support different types of applications, and cryptocurrencies are just one of those application.

What should be of most interest to psychologists and researchers alongside the content of the data is that much of this data is open and permissionless. Anyone can participate in the network, perform transactions, and store or retrieve information on the blockchain anonymously and without restrictions.

Public, permissionless, blockchains (like the Bitcoin and Ethereum networks) offer huge potential for researchers, with billions of data points representing hundreds of millions of people's real-time behaviour. Blockchain is a treasure trove of open-source behavioural data.

Since 2009, nearly 1 billion bitcoin transactions have been completed. These are only a small fraction of the total amount of data recorded on blockchains, but these data are open and can be accessed and analysed by any researcher. The data can be used to measure individual behaviour which are linked to digital wallets (like user accounts), or to conduct large-scale, longitudinal studies across time. 

Now many other blockchain networks work like global computers. Anyone can connect to these computers, and anyone can create their own computer programs that run on these networks. The Ethereum network, for example, has processed, on average, over 1 million transactions every day since 2019. These are not just financial transactions, they also represent activity in online games, trading and transferring digital goods like art, music, books, interactions on social networks, records of attendance at conferences and events, along with voting, governance, and social coordination activities. 

Blockchain networks are also creating comprehensive digital profiles of individuals, linked to digital wallets, within an interconnected network of open but pseudo-anonymous data. Digital behaviour creates complex data which represent meaningful psychological variables like personality, recreational and consumer preferences, social interactions and group affiliations. The European Union uses blockchain to store data about educational and professional qualifications and are rolling out a program to use blockchain-based verifiable credentials to confirm educational and professional qualifications across borders. The EU are also expanding their blockchain infrastructure to track physical goods, to digitise IDs and public services data as part of the European Blockchain Services Infrastructure. 

A window into previously inaccessible behaviour

Bitcoin has a huge volume of transactions which are continuous records of both individual and systems-level behaviour. Researchers are beginning to use that wealth of historical, publicly available data for analysing digital behaviour. 

For example, Bracci and colleagues (2021) analysed 245 million transactions that took place between 2010 and 2021 (worth USD$25 billion) to measure behaviour across 28 online marketplaces. They measured financial activity in networks of buyers and sellers in both regulated and illicit (dark web) marketplaces. In the same way as any business has a bank account and payment details, a dark web marketplace will have a digital address to send cryptocurrency. But on a public blockchain network, like Bitcoin, all the data are public. This means it is possible to see all the addresses sending cryptocurrencies to a dark web marketplace (buyers) and all the addresses receiving cryptocurrencies from the dark web marketplace's cryptocurrency address (sellers).

Many people think cryptocurrency transactions are anonymous. However, most cryptocurrency transactions, like those of Bitcoin, are psedoanonymous at best. Recent estimates (Dearden & Tucker, 2023) suggest that over 90 per cent of cryptocurrency transactions are identifiable. Every transaction is recorded in a network of connections, and researchers like dos Reis and colleagues (2024) have shown they can identify the major actors across marketplaces, to map the behaviour of the most prolific buyers and sellers.

Such research has looked at network-level variables of marketplaces, but individual differences researchers could extract subsets of data from this network. For example, a researcher could investigate all the users that made a purchase from a particular marketplace or seller. It is then possible to investigate the digital behaviour of all the users that have transacted with those marketplaces or sellers. Other researchers have used machine-learning techniques to analyse blockchain activities to develop behavioural profiles of users based on historical activity (e.g. Valadere et al., 2023).

An interesting feature of Bracci and colleague's research is that network-level data made it possible to identify large clusters of transactions that could be linked to specific users and specific marketplace activity. This makes it possible to measure aspects of digital and financial behaviour that would normally be difficult or impossible for most researchers to access: especially in large numbers. 

Their findings also provided interesting insight into the similarities between dark web marketplaces and typical, regulated and legal online marketplaces. The financial and social structure of the activity was remarkably similar on regulated online marketplaces that are subject to regular financial and consumer protection rules, and across dark web marketplaces that relied on reputation and trust with no legal enforcement mechanisms. In both e-commerce networks a small number of well-known and reputable sellers tend to dominate the market where marketplaces are heavily influenced by 'preferential attachment': buyers tend to return to the location of their most recent successful transaction. This would seem to indicate there are fundamental social and psychological processes at play: but these digital frontiers are, as of yet, still largely unexplored by psychologists.

Risky digital financial behaviour

Blockchain, and cryptocurrency, often receives attention for spectacular collapses. Yet unlike when a private company or a government department has a meltdown, blockchain collapses happen in public. This also means that most or all associated data are publicly available.

For instance, in 2022, a notable case involved a piece of blockchain software (called Anchor) designed to automate savings and loans accounts with cryptocurrency (Terra and Luna), allowing for the deposit, withdrawal, or borrowing of digital tokens. This software experienced one of the most rapid ascents and dramatic falls in the history of cryptocurrency so far. At its peak in May 2022, the software was associated with approximately $50 billion in valuation. By June, its value had plummeted to nearly zero.

The Terra/Luna debacle provided researchers with an opportunity to examine large-scale social activity, analysing 228 million transactions from 3.7 million addresses across 19 months from October 2020 to May 2022). Researchers Lui, Makarov, and Schoar published their findings for the National Bureau of Economic Research in the United States. They found that the behavioral patterns mirrored those of a traditional bank run.

The researchers could model the exact trajectory of the bank run. They identified specific social signals that precipitated the collapse. Large withdrawals from a few cryptocurrency wallets were seen as a social signal: 'common public signals can act as a coordination mechanism and trigger a run by investors. In the context of cryptocurrencies, where the actions of agents on the blockchain are publicly observable and agents can monitor and react to each other's actions, the actions themselves can serve as a coordination mechanism.'

Instead of a bank run in the physical world, where people use visual and social clues from physical spaces as social signals and coordination mechanism (people physically lining up to withdraw their money), blockchain users can observe financial transactions in real-time online. Large withdrawals or notable changes in behaviour from those with significant amounts of financial or social capital can trigger a wave of activity that is rapidly imitated. Lui and colleagues (2023) found that in this case, the digital bank run unfolded in generally the same pattern as bank runs have occurred throughout history.

The researchers also noted that open data does not necessarily lead to better decision-making by the majority. Although everyone in the network had access to the same publicly available information, larger and more experienced investors interpreted the coordination mechanisms differently. 'These results underscore the fact that observability and free access do not, by themselves, level the playing field for investors if there are significant differences in their ability to process and interpret information.' This poses interesting questions for psychologists about how different people process and interpret information and social signals in massive, open digital systems. Even when people all have access to the same data, they use that data to make different decisions. 

Blockchain data and opportunities for researchers

It's difficult to dismiss blockchain networks as a fad. They have been around for 15 years now, and their data has been public since inception. They offer a window into the digital behaviour of hundreds of millions of people, connected around the world through decentralised digital network activity. That trajectory of user numbers is moving swiftly upward.

Blockchain also creates opportunities with open data in ways that previous waves of technology never did. For example, the social networking applications that emerged in the early 2000s and exploded in size in the 2010s have received much attention by researchers. Yet most of the data collected and algorithms created by social media companies is proprietary and closely guarded. Most blockchain data is open and permissionless, but requires a degree of technical skill to access, use and analyse.

So far, most of the academic research on blockchain networks has been conducted by mathematicians, computer scientists and economists. The question is: when will psychologists take notice of this ever-expanding mountain of digital behavioural data that is freely available?

Ian MacRae is director of High Potential Psychology, a psychometrics and research company. He is an award-winning author of books including High Potential, Dark Social, and his latest book is Web of Value: Understanding blockchain and web3's intersection of technology, psychology and business. He is currently serving as a member of the British Psychological Society's Division of Occupational Psychology committee.

References

Bracci, A., Boehnke J.,  El Bahrawy, A., Perra, N., Teytelboym, A., & Baronchlli, A. (2022). Macrosopic properties of buyer-seller networks in online marketplaces. PNAS Nexus, 1(4).

dos Reis, E.F., Teytelboym, A., ElBahrawy, A. et al. (2024). Identifying key players in dark web marketplaces through Bitcoin transaction networks. Scientific Reports 14, 2385.

European Commission (n.d.). European Blockchain Services Infrastructure.

European Union Agency for Law Enforcement Cooperation (Europol). (2022). Europol Spotlight: Cryptocurrencies: Tracing the evolution of criminal finance.

Liu, J., Makarov, I., Schoar, A. (2023). Anatomy of a run: The Terra Luna Crash. National Bureau of Economic Research. Working Paper 31160.

Nadini, M., Alessandretti, L., Di Giacinto, F., Martino, M., Aiello, L. M., & Baronchelli, A. (2021). Mapping the NFT revolution: market trends, trade networks, and visual features. Scientific Reports, 11, 20902.

Dearden, T. E., & Tucker, Samantha. (2023). Follow the Money: Analyzing Darknet Activity Using Cryptocurrency and the Bitcoin Blockchain. Journal of Contemporary Criminal Justice, 39(2). 10.1177/10439862231157521.

Valadares, J.A. Villela, S. M., Bernardino, H. S., Gonçalves, G. D., Vieira, A. B. (2023). Mapping user behaviors to identify professional accounts in Ethereum using semi-supervised learning, Expert Systems with Applications,229B, 120438.