19.6 The Random Walk: The Simplest Stochastic Model

19.6 The Random Walk: The Simplest Stochastic Model

So far, we have introduced stochastic processes as mathematical models that generate an ensemble of possible trajectories rather than a single deterministic prediction.

The next question is naturally:

How does a stochastic trajectory actually arise?

To answer this question, we begin with one of the simplest stochastic models in mathematics—the random walk.

Although remarkably simple, the random walk forms the foundation of many biological models, including molecular diffusion, intracellular transport, bacterial chemotaxis, animal movement, and population genetics.

Like the logistic map in the previous chapter, the random walk demonstrates how surprisingly rich behavior can emerge from an extremely simple rule.

A simple experiment

Imagine a molecule moving along a one-dimensional line.

At every time step, the molecule has two possible choices.

It may

• move one step to the left, or
• move one step to the right.

Both movements are equally likely.

Suppose the molecule starts at position

$$x=0.$$

At the first step, it moves either to

$$x=-1$$

or

$$x=+1.$$

The same rule is then applied again.

Each new position depends only on the current position and the outcome of the next random step.

The trajectory therefore evolves through a sequence of random decisions.

One realization

If we perform this experiment once, we obtain a single trajectory.

For example,

0 → 1 → 2 → 1 → 2 → 3 → 2 → 1 ...

Running the experiment again produces a completely different path,

0 → -1 → 0 → -1 → -2 → -1 → 0 ...

Neither trajectory is more correct than the other.

Both are equally valid realizations of the same stochastic process.

This illustrates an important principle.

Unlike deterministic models, stochastic models do not predict one unique trajectory.

Instead, every simulation generates one possible history of the system.

Many realizations

A single random walk reveals very little about the underlying process.

The true behavior only becomes apparent when we repeat the experiment many times.

Some trajectories wander far to the left.

Others drift to the right.

Many remain close to the starting position.

Together, these trajectories form an ensemble.

The ensemble is the true object of interest.

It reveals the probability distribution of particle positions after a given number of steps.

Emergent regularity

At first glance, the trajectories appear completely random.

Surprisingly, however, the ensemble exhibits a highly regular pattern.

Although the individual trajectories differ greatly, the overall probability distribution becomes increasingly smooth.

After many repeated experiments, the particle positions form a characteristic bell-shaped distribution centered around the starting point.

This observation illustrates an important feature of stochastic systems.

Random behavior at the level of individual trajectories can generate highly predictable statistical behavior at the population level.

From trajectories to distributions

The random walk demonstrates a profound conceptual shift.

In deterministic modelling, a trajectory is usually the final prediction.

In stochastic modelling, an individual trajectory is only one realization.

The real prediction is the distribution obtained from many realizations.

This distinction lies at the heart of stochastic modelling.

The objective is no longer to predict exactly where one molecule will be after one hundred steps.

Instead, we ask:

• Where are molecules most likely to be?
• How broad is the distribution?
• How does the distribution change over time?

These questions naturally lead from individual trajectories to population-level behavior.

Biological examples

Although the random walk is mathematically simple, it describes many important biological processes.

Examples include

• diffusion of metabolites within cells,
• movement of signaling molecules,
• Brownian motion of proteins,
• migration of microorganisms,
• dispersal of seeds,
• and exploratory movements of animals searching for food.

In each case, individual movements are partly random.

Nevertheless, the collective behavior follows predictable statistical laws.

Looking ahead

The simple random walk assumes that movement to the left and right is equally likely.

Many biological systems behave differently.

Cells may move toward nutrients.

Animals may follow environmental gradients.

Molecules may experience directed transport.

To describe such behavior, we extend the random walk by introducing biased probabilities, allowing random motion and directed movement to coexist.

Key concepts

• A random walk consists of a sequence of random steps.
• Each simulation represents one realization of the stochastic process.
• Many realizations form an ensemble.
• Individual trajectories are unpredictable, but their statistical properties are highly predictable.
• Stochastic models focus on probability distributions rather than individual trajectories.
• Random walks provide a minimal model for many biological transport processes.

Summary

The random walk is one of the simplest stochastic models and provides a foundation for understanding biological randomness. Each realization follows a different trajectory because every step is determined probabilistically. Although individual trajectories are unpredictable, repeated simulations reveal highly regular probability distributions. This illustrates the central idea of stochastic modelling: predicting ensembles and distributions rather than single system trajectories.

Self-check questions

1. What rule defines a simple random walk?
2. Why do repeated simulations produce different trajectories?
3. What is meant by a realization of a random walk?
4. Why is an ensemble more informative than a single trajectory?
5. Why does a random walk produce a smooth probability distribution despite its random steps?
6. Which biological processes can be approximated by random walks?