Problem 35 - Calculus

Question

Policy makers are interested in modeling the spread of information through a population. For example, agricultural ministries use models to understand the spread of technical innovations or new seed types through their countries. Two models, based on how the information is spread, follow. Assume the population is of a constant size \(M\) (a) If the information is spread by mass media (TV, radio, newspapers), the rate at which information is spread is believed to be proportional to the number of people not having the information at that time. Write a differential equation for the number of people having the information by time \(t .\) Sketch a solution assuming that no one (except the mass media) has the information initially. (b) If the information is spread by word of mouth, the rate of spread of information is believed to be proportional to the product of the number of people who know and the number who don't. Write a differential equation for the number of people having the information by time \(t .\) Sketch the solution for the cases in which (i) No one \(\quad\) (ii) \(5 \%\) of the population (iii) \(75 \%\) of the population knows initially. In each case, when is the information spreading fastest?

Step-by-Step Solution

Answer

The differential equations are: media: \( \frac{dN}{dt} = k(M-N(t)) \), word: \( \frac{dN}{dt} = kN(t)(M-N(t)) \). Spread peaks when \( N(t) = M/2 \).

Step 1: Understanding the Problem

We are asked to write differential equations modeling information spread in two different scenarios - by mass media and by word of mouth. We need to represent the number of informed people as a function of time in both cases, then sketch probable trends of how information spreads initially based on different starting conditions.

Step 1: Formulate the Differential Equation for Mass Media Spread

When information is spread by mass media, the rate is proportional to the number of people yet to be informed. Let \( N(t) \) represent the number of informed people at time \( t \). The remaining uninformed population will then be \( M - N(t) \). The differential equation is:\[ \frac{dN}{dt} = k (M - N(t)) \]Where \( k \) is the proportionality constant.

Step 2: Sketch the Solution for Mass Media Scenario

Initially, when no one knows the information (i.e., \( N(0) = 0 \)), the spread will be rapid as there are many uninformed people. The rate reduces over time as fewer are left to inform. This is a typical first-order linear differential equation resulting in exponential growth that approaches saturation \( N(t) \to M \) over time.

Step 3: Formulate the Differential Equation for Word of Mouth Spread

For word of mouth, spread is proportional to interactions between informed and uninformed people. Use \( N(t) \) for informed and \( M - N(t) \) for uninformed. The equation is:\[ \frac{dN}{dt} = k N(t) (M - N(t)) \]This logistic equation reflects growth that is initially exponential but slows as saturation is approached.

Step 4: Analyze and Sketch Word of Mouth Solutions for Different Initial Conditions

(i) If initially no one knows, \( N(0) = 0 \), so initially, there is no spread until a first piece of information passes.(ii) If \( N(0) = 0.05M \), it starts with a small known base; spread accelerates as more individuals become informed—peaks at \( N = M/2 \).(iii) If \( N(0) = 0.75M \), most know initially; spread is sluggish as time progresses, peaking when about half are uninformed.

Step 5: Determine Fastest Spread Periods

For word of mouth, spread occurs fastest when \( N(t) = M/2 \), where the product of informed and uninformed populations is maximized.

Key Concepts

Information Spread Modeling

When we talk about information spread modeling, we're looking at how information flows through a community or population over time. Different models can be used to predict and understand this flow, and they often involve differential equations. These equations express the rate of change in the number of informed people at any given time. The goal is to capture how quickly or slowly information spreads under different conditions and through different channels.

Understanding how information spreads is essential for many fields. For example, it can be applied in public health to understand how awareness about a disease circulates. It also helps policymakers maximize the reach of critical information in a population.

Two primary methods of spreading information are often modeled: via mass media, like TV and newspapers, and through word of mouth, where people directly share information with one another. Each of these methods influences the spread rate differently, forming the basis of many information spread models.

Mass Media Model

The mass media model is one of the simplest ways to understand how information spreads in a population. Here, the spread rate is directly linked to how many people still need to receive the information. The differential equation for this scenario is:\[ \frac{dN}{dt} = k (M - N(t)) \]Where:

• \( N(t) \) represents the number of informed people at time \( t \).
• \( M - N(t) \) is the number of uninformed people.
• \( k \) is a constant that represents the efficiency of mass media.

This model assumes the population size \( M \) is constant, and the initial condition is often \( N(0) = 0 \), meaning no one knows at the start except the media.

The spread under this model starts quickly when the whole population is uninformed and slows down as more people get informed. This exponential growth eventually levels off, approaching a saturation point where almost everyone is informed.

Word of Mouth Model

The word of mouth model offers a more complex picture of how information spreads. It assumes that information is passed from person to person, meaning the rate of spread depends on both informed and uninformed people interacting. The governing equation for this model is:\[ \frac{dN}{dt} = k N(t) (M - N(t)) \]This logistic growth model shows that the spread rate increases as more people know, reaching a peak when half the population is informed. After this point, the spread rate decreases until saturation is reached.

This model is more dynamic as it accounts for interactions within the population. It considers various starting points:

• When no one knows initially, \( N(0) = 0 \), the spread doesn't begin until someone starts the chain of information.
• If \( 5\% \) of the population knows, the spread accelerates quickly, peaking when half the population is informed.
• With \( 75\% \) already informed, the spread is slower, as fewer people are left uninformed.

The spread occurs fastest when the informed and uninformed populations are evenly split.

Logistic Growth

Logistic growth is a crucial concept in understanding population dynamics and information spread. In these contexts, it describes how a quantity increases quickly at first but then slows as it approaches a maximum limit.The logistic growth model is often expressed with the differential equation:\[ \frac{dN}{dt} = k N(t) (1 - \frac{N(t)}{M}) \]This equation incorporates the idea that growth is initially exponential but gradually slows as resources (or uninformed people, in the case of information spread) become limited.

In information spread, the logistic model helps illustrate how the spread rate is tied to the number of people yet to be informed. Initially fast, it decreases over time and approaches a plateau.

Understanding logistic growth is vital because many real-world phenomena, from animal populations to market penetration of a new product, follow this pattern. It gives a realistic view of growth dynamics, highlighting how initial rapid growth can sustainably lead to eventual stability.