What can Pac-Man, Space Invaders, and Mortal Kombat teach us about cultural evolution?

In shopping malls of the 1970s, arcades were where people played digital games together, huddled around custom-built cabinets. These coin-operated arcade video games, designed for short, replayable sessions, enabled the public to interact with computers.

As a social-technical phenomenon, arcade games were part of the origins of early computer-human interaction, long before smartphones or chatGPT.  In a recent study published in Evolutionary Human Sciences, we looked at this unique example of early video game evolution. We investigated the rise and fall of more than 7,000 arcade video games—treating them not just as entertainment artifacts, but as components of a larger cultural and technological ecosystem. We documented how arcade games emerged and diversified, and their evolution sometimes collapsed.

Early arcade games were evolving systems. Drawing on detailed, structured metadata—including genre classifications, developer collaborations, and technical specifications like ROM size and CPU speed—we analyzed their innovation across genres and across time.

Our dataset is derived from MAME (Multiple Arcade Machine Emulator)—the open-source archive of arcade game hardware systems. MAME not only keeps classic games playable; it also encodes rich metadata on each machine’s technical configuration, release context, and design lineage.

Instead of focusing on individual works or creators, like most studies of innovation, we approached the entire dataset as a complex system. This way, we could detect broader feedback loops and evolutionary dynamics—showing how genres rise, stall, and sometimes collapse not because of content alone, but because of structural conditions.

Genres as Cultural-Ecological Niches

In the arcade world, genres like Maze, Shooter, Fighter, or Driving have distinctive gameplay mechanics and technical demands. Never static, these genres evolved like cultural-ecological niches: changing, branching, and sometimes dying out.

We visualized this evolution using an ecological diagram called a morphospace (see Figure). This is a conceptual design space defined by each game’s hardware traits. In the Figure, the horizontal axis shows ROM size (roughly, the game’s audiovisual and logical complexity). The vertical axis shows CPU speed (processing power). As games accumulated over time, genres moved through this space in distinct patterns.

Figure: Left: The morphospace of arcade games shows how genres explore the landscape of hardware complexity. Games like Pac-Man, Gradius, and Mortal Kombat represent distinct trajectories through CPU and ROM space. Right: A branching tree of subgenres reveals diversification, stalling, and collapse over time. Some genres persist as “living fossils”; others fragment or vanish after saturation.

Our morphospace represents game genres like evolving species rather than individual inventions. Some genres, like Fighter (think Mortal Kombat) or Driving games (think Out Run), pushed into unexplored regions of the morphospace—leveraging hardware advances to introduce novel gameplay experiences. Others, like Maze games (think Pac-Man), became trapped in crowded regions of the morphospace, spawning waves of similar clones without much novelty—something we call the “dilution of expertise.”

We quantified these patterns using two types of scaling laws. One measured how technical traits—like memory and processing—scaled together over time (allometric scaling). The other, adapted from Heap’s Law, measured how many new traits each genre introduced as it grew.

The result? Genres that fostered collaboration and recombination maintained higher innovation rates and greater trait diversity. Imitation-dominated genres experienced a halt in their creative evolution.

Genres Collapsing

Our concept of evolutionary “collapse” is not about crashing popularity or cultural irrelevance. Instead, it is about the collapse of its innovation and evolutionary failure. The evolutionary collapse of an arcade game genre occurred when bootlegs and clones began to outnumber collaborations. The genre became stale and no longer generated novelty. The genre didn’t necessarily disappear. Some collapsed genres—like Maze—remained popular and visible for years. The games could remain popular, but like living fossils, their genre was creatively stagnant. Its evolutionary future has closed.

A dramatic example of this is the 1983 arcade crash, when oversaturation of low-quality and derivative games contributed to a market-wide downturn. But many collapses we observed were quieter—gradual, systemic, and rooted in over-replication rather than economic failure.

Maze games, for instance, once sparked a flurry of design variation after Pac-Man. But as bootlegs proliferated and collaboration declined, novelty dried up. By the late 1980s, the genre had effectively stalled.

In contrast, Fighter games (like Mortal Kombat) and Driving games (like Out Run) showed greater resilience. These genres encouraged collaborative production, sometimes involving multiple studios or companies. As a result, they maintained a high level of trait diversity—developing new mechanics, perspectives, and hardware integrations.

Lessons Beyond the Arcade

The evolution of arcade games reveals how creative systems evolve and how they sometimes get stuck. Whether in music, literature, open-source software, or science, the mode of production matters.  More than individual creative design, innovation requires a sociotechnical ecosystem—other games, genres and their players—that promotes novelty and recombines it into new forms.

We call it cultural macroevolution. Like languages or tools, game genres evolve under both technical and social constraints. Using metadata—collaboration tags, trait profiles, and genre trees—we can track how creative ecosystems shift over time and what leads them to thrive or collapse.

In this view, innovation is a system-level process that relies on long-term collaboration, technological flexibility, and recombinatory potential. And when these conditions deteriorate, even the most iconic cultural forms could face evolutionary stagnation. The revealed hat-like pattern of innovation, saturation in diversity, decline, and persistence of some forms in a long tail of ‘living fossils’ is universal for biological and cultural evolution, demonstrating deep similarities in seemingly disparate complex systems.

For more on this topic, read The cultural macroevolution of arcade video games: innovation, collaboration, and collapse, published in Evolutionary Human Sciences.

Authors Biography

Sergi Valverde is a complex systems scientist at the Institute of Evolutionary Biology (CSIC-UPF) in Barcelona and an external DySoC member (University of Tennessee).