Drawing, printmaking, and surrealist paintings depicting human figures, animals, and nature were Jackson Pollock’s first forays into abstract art.

It has been suggested that Pollock’s period of experimentation, followed by the use of his new drip technique, set him up for a “hot streak” of high-impact paintings, according to new research conducted by Penn State experts. Between 1947 and 1950, Jackson Pollock produced all of his drippy, splattered masterpieces, which have made him famous to this day.

Artificial intelligence helped academics uncover that this trend is not a random one but rather a mystical recipe for the success of artists, film directors and scientists. After years of inquiry (studying various styles and topics), they discovered that hot streaks are precisely related to the years of exploitation that had immediately followed (focusing on a narrow area to develop deep expertise).

Nature Communications reported the findings last month.

A better knowledge of what causes a hot streak can help institutions design settings that encourage and support their members’ success.

It’s not just exploration and exploitation that are linked to hot streaks. “It’s the combination of them,” Dashun Wang, a professor at Northwestern University’s Kellogg School of Management, who led the study, said. Because there’s always the possibility that you won’t find anything useful, exploration is seen as a risk. Exploitation, on the other hand, is commonly seen as a cautious approach. It’s possible that using the same type of work over and over again may inhibit innovation. Even more curiously, “exploration and exploitation appear to have consistent associations with the onset of hot streaks.”

According to Lu Liu, a doctorate candidate in information sciences and technology at Penn State and the paper’s first author, the study reveals one of the first empirical regularities underlying the commencement of a hot streak across diverse disciplines.

This study challenges the conventional wisdom that hot streaks and individual creativity are random events, according to Liu. “However, the paper suggests that the start of a hot streak isn’t completely random.” People, on the other hand, tend to look at a variety of possibilities, weigh their options, and then take advantage of the most promising ones.”

“The existence of hot streaks has been much debated,” said C. Lee Giles, David Reese Professor of Information Sciences and Technology at Penn State and study co-author. Their existence has been shown by this study.

With the help of Penn State professor C. Lee Giles and Northwestern researchers Wang and Xu, they published a Nature study in 2018 describing the hot streaks of creative, cultural, and scientific careers. To find out what causes these hot streaks, the researchers first had to prove they happened. When Wang visited the Van Gogh Museum in Amsterdam, he discovered a clue.

The Starry Night, Sunflowers, and Bedroom in Arles were all painted during Van Gogh’s artistic breakthrough between 1888 and 1890. His earlier work, on the other hand, was more realistic and less impressionistic. Instead of his more well-known use of bright, sweeping hues, he preferred using gloomy earth tones.

Before 1888, Wang explained, “his output was all over the place.” In contrast to the work he produced while on a hot streak, the collection was full of still-life paintings, pencil sketches, and portraits.

This study used deep learning algorithms and network science to construct computational tools and then applied these methods to enormous datasets tracking the career outputs of artists, film directors, and scientists. The results were promising.

Images of art from museums and galleries were mined for data on 2,128 artists, including Jackson Pollock and Vincent Van Gogh, using algorithms for image recognition. The Internet Movie Database (IMDb) was used to compile director data sets, with a total of 79,000 films and 4,337 directors represented. Data from the Web of Science and Google Scholar were used to examine the careers of 20,040 researchers.

With the help of other researchers, Liu and her colleagues were able to identify hot streaks in each individual’s career based on the effect of the works they produced. Afterwards, they made a connection between the occurrence of hot streaks and the individual’s creative trajectory. The researchers studied the careers of people four years before and after the hot streak to see how their work altered around the start of a hot streak.

The scientists discovered that the likelihood of a hot streak was dramatically lowered when exploration was not followed by exploitation. In the same way, exploitation without exploration does not ensure a winning streak. In contrast, researchers saw a considerable rise in the likelihood of a hot streak if exploration was closely followed by exploitation.

Liu emphasized the importance of a long-term impact. When it comes to spotting, developing, and fostering new talent, this is especially important, given the current state of exploration-exploitation dynamics, which includes everything from publication pressure to tenure review.

According to study co-author Jillian Chown, an associate professor of management and organizations at Kellogg School, “This knowledge can help both individuals and organizations understand the different types of activities to engage in — such as exploring new domains or exploiting existing knowledge and competencies — and the optimal sequence to use in order to achieve the most significant impact.”

About five years is the average length of a hot streak There are no longer any patterns of exploration or exploitation observed at this point, according to the researchers.