There is no dearth of financial information to access, evaluate, and exchange on social media. So great are the quantities of information that it is investor attention that has become a limited resource because of time and processing capacity constraints. The authors shed additional light on the cognitive processes that underpin investors’ attention, using observations from Twitter exchanges as part of their experiment.

Using investors’ social media activity as their test subject, the authors critically explore the mechanisms of investor attention, which operates according to cognitive niches or evolved processes to interact with and process information. The time frames under which investors use their attention to process information break with the implications of the efficient market hypothesis (EMH), which assumes investors’ uniform ability to focus attention on all available information in a timely fashion. Investor attention is fickle according to the context in which it operates and the skills and experience of the investor. The interaction among these factors accounts for the nature of attention among investors, which academia has traditionally deemed a mystery within which operates a single static—rather than dynamic—mental bias.

A plethora of information and time pressure constrain investors’ focus. Attention proceeds along varying investor time scales and may be subject to “contaminants” in the forms of the search for validation or social proof (that is, imitating the behavior of others in order to conform to correct behavior). This contagion is elevated in the presence of heightened volatility or stronger return momentum. The authors’ simulation of an “attentional market” of stocks seems to confirm a stronger and more robust correlation between quantities of investor attention and trading volumes for those equities with active investor focus.

Students of behavioral finance as well as traders and portfolio managers would benefit from a deeper understanding of how investor attention operates and interacts with security trading patterns. To the extent that investor attention drives variability in stock returns, risk managers have much to learn about this source of volatility.

An initial review of the relevant literature on investor attention highlights its struggle to explain the predictability of asset returns and its focus on behavioral biases. Previous researchers have not focused on how investor attention can be selective. The authors explore variations of investors’ attention on social media—specifically, Twitter.

Twitter activity gives the authors the opportunity to observe investor attention in real time. Specifically, they consider individuals’ Twitter activity on the well-known investor community StockTwits.com over a trading period longer than 350 trading days—from 17 May 2011 to 3 October 2012. For each tweet, the authors identify the author, the ticker symbol of the financial asset, and a time stamp with resolution in seconds. They observe the frequency of tweets for specific securities and characterize the investors under observation as novice, intermediate, or professional in terms of expertise and as technical or fundamental in terms of approach. Analyzing individual data that are time stamped in seconds provides much greater analytical precision than such other methods as analyzing weekly aggregates of Google search activity for a given stock.

The authors parse the data for plausibility of investor attention assumed in the EMH model versus real-world experience. They evaluate how cognitive niches influence distributions of data and then how the use of these niches is a function of context (e.g., uncertainty). The authors observe that no single mechanism regulates investor attention. Rather, such environmental circumstances as uncertainty, along with investor skill, experience, and methodology, drive the function of investor attention. The role of social proof—that is, how crowd behavior draws investor attention—can be both significant and powerful. To this end, the authors design an experiment to consider investor reaction to stock price jumps, controlling for time periods of activity. Through regression analyses, they quantify the interaction of the effects of social proof and environmental cues. Investors’ behavior can be adaptive; their use of cognitive niches implies that there are degrees of investor attention elasticity—or, the extent to which it responds to trading volume—in well-followed versus less-followed stocks.

Stocks with higher volatility or price momentum seem to better capture investor attention. Because some investors have broader time scales for engaging with a stock, the periods of volatility can last for longer periods of time. The lower the frequency of news on a stock, the greater the impact of a specific report on investor attention to the stock. An increase in investor attention coincides with increases in trading volume and precedes unusual trading volume by more than one week. These metrics may be predictive of stock returns three days in advance. When investors are paying greater attention to a stock and a future event, volatility rises preceding a price increase, whereas volatility reaction is much more muted in stocks with passive investor attention.

The selectivity of investor attention is a function of the interplay of evolutionary biology and situational factors—including an information glut, time pressure, and limitations on cognitive processing capacity. The authors aspire to a greater understanding of how human investing intelligence operates in the modern, fast-paced information world of social media. Twitter is an ideal environment to test their hypotheses. How the predictive powers of investor attention affect market risk is a possible topic for further study.