Part 3: Decision-Making Models in Service_Design

In Part 3 of this lesson, we delve into the Decision-Making Models that are integral to understanding and influencing user behavior in service design. We will explore key theories such as Prospect Theory, the concept of Utility Functions, and the application of Game Theory in service interactions. By comprehensively understanding these models, service designers can create more effective strategies that align with the ways users make decisions, ultimately enhancing user experience and service effectiveness.

9.14 Introduction to Decision-Making Models

9.14.1 The Role of Decision-Making Models in Behavioral Economics

Decision-making models in behavioral economics seek to explain how individuals make choices under conditions of uncertainty and risk. Unlike traditional economic models that assume rationality, these models account for psychological factors and cognitive biases that influence decisions.

Importance in Service_Design:

• Predicting User Behavior: Helps in anticipating how users will respond to different service offerings.

• Designing Incentives: Guides the creation of incentives that align with user motivations.

• Enhancing User Experience: Improves the design of services by aligning with natural decision-making processes.

9.14.2 Overview of Key Decision-Making Models

• Expected Utility Theory: Traditional model based on rational choice.

• Prospect Theory: Accounts for how people perceive gains and losses.

• Game Theory: Analyzes strategic interactions among rational decision-makers.

• Heuristics and Biases: Simplifies decision-making processes but can lead to systematic errors.

9.15 Expected Utility Theory and Its Limitations

9.15.1 Expected Utility Theory

Definition:

Expected Utility Theory (EUT) posits that individuals make decisions by considering the expected outcomes and selecting the option that maximizes their expected utility.

Mathematical Representation:

Let $U$ be the utility function, $x_i$ be possible outcomes, and $p_i$ be the probabilities of each outcome.

Assumptions:

• Rationality: Individuals are rational and seek to maximize utility.

• Transitivity: Preferences are consistent (if A > B and B > C, then A > C).

• Independence: Preferences between options are independent of irrelevant alternatives.

9.15.2 Limitations of Expected Utility Theory

• Descriptive Inaccuracy: Fails to predict actual human behavior in many situations.

• Neglects Cognitive Biases: Does not account for biases like loss aversion or framing effects.

• Complex Calculations: Assumes individuals can compute and compare expected utilities accurately.

9.16 Prospect Theory

9.16.1 Introduction to Prospect Theory

Originators:

Daniel Kahneman and Amos Tversky introduced Prospect Theory in 1979 as an alternative to Expected Utility Theory.

Core Concept:

Prospect Theory describes how people make decisions in situations involving risk and uncertainty, emphasizing that people value gains and losses differently, leading to irrational decision-making.

9.16.2 Key Components of Prospect Theory

1. Reference Point Dependence:

   • Decisions are made based on potential gains or losses relative to a reference point (often the status quo).

2. Loss Aversion:

   • Losses have a greater emotional impact than an equivalent amount of gains.

   • Typically, losses are felt about 2 to 2.5 times more intensely than gains.

3. Diminishing Sensitivity:

   • The subjective difference between $100 and$200 is perceived as greater than between $1,100 and$1,200.

   • Applies to both gains and losses.

4. Probability Weighting:

   • People tend to overweigh small probabilities and underweigh large probabilities.

   • Leads to risk-seeking in losses and risk-averse behavior in gains.

9.16.3 The Value Function

Characteristics:

• S-Shaped Curve: Concave for gains (implying risk aversion) and convex for losses (implying risk-seeking).

• Steeper for Losses: Reflects loss aversion.

Mathematical Representation:

The value function $v(x)$ can be represented as:

Where:

• $x$ = change in wealth relative to the reference point.

• $\alpha, \beta$ = parameters reflecting diminishing sensitivity (usually between 0 and 1).

• $\lambda$ = coefficient of loss aversion (usually $\lambda > 1$).

Graphical Representation:

• The value function is defined over gains and losses rather than total wealth.

• The origin represents the reference point.

• The function is steeper in the domain of losses.

9.16.4 Probability Weighting Function

Definition:

Individuals do not perceive probabilities linearly; instead, they use a weighting function $\pi(p)$.

Properties:

• Overweighting Small Probabilities: $\pi(p) p$ when $p$ is small.

• Underweighting Large Probabilities: $\pi(p) < p$ when $p$ is large.

Commonly Used Function:

One parametric form for the weighting function is:

Where $\gamma$ reflects the curvature of the weighting function.

9.16.5 Calculating Prospect Theory Value

Example:

Consider a choice between:

• Option A: 50% chance to win $1,000.

• Option B: $450 guaranteed.

Traditional Expected Utility:

• EU of A: $0.5 \times U(1,000) + 0.5 \times U(0)$

• EU of B: $U(450)$

Prospect Theory Calculation:

1. Determine Reference Point:

   • Assume current wealth is the reference point.

2. Compute Gains/Losses:

   • Option A: Gain of $1,000 with$p = 0.5$, else no gain.

   • Option B: Gain of $450 guaranteed.

3. Apply Value Function:

   • Use $v(x) = x^\alpha$ for gains.

   • Assume $\alpha = 0.88$.

4. Apply Probability Weighting:

   • Use $\pi(p)$ to adjust probabilities.

   • Assume $\gamma = 0.61$.

5. Calculate Weighted Values:

   • Option A Value:

• Option B Value:

$V_B = v(450) = (450)^{0.88}$

• Compute $\pi(0.5)$ Using the Weighting Function.

6. Compare $V_A$ and $V_B$:

   • If $V_A V_B$, prefer Option A; else, prefer Option B.

Interpretation:

Prospect Theory may predict a preference for the certain gain (Option B) due to the overweighting of certain outcomes and loss aversion.

9.17 Applications of Prospect Theory in Service_Design

9.17.1 Pricing Strategies

• Partitioned Pricing:

  • Breaking down prices into components (e.g., base price + fees).

  • Users may perceive the base price as a gain and fees as smaller losses.

• Bundling Services:

  • Combining multiple services at a single price can be perceived as a larger gain.

9.17.2 Framing Offers

• Gain-Framed Messaging:

  • Emphasize the benefits (e.g., “Save $200 by enrolling now”).

• Loss-Framed Messaging:

  • Highlight what users stand to lose (e.g., “Don’t miss out on a $200 saving”).

9.17.3 Managing Risk Perception

• Insurance Services:

  • Emphasize protection against potential losses.

• Investment Services:

  • Present potential gains and losses, considering loss aversion tendencies.

9.17.4 Designing Incentives

• Loyalty Programs:

  • Offer points or rewards that accumulate, framing gains.

• Penalty Fees:

  • Late fees can leverage loss aversion to encourage timely payments.

9.18 Utility Functions in Decision-Making

9.18.1 Definition of Utility Functions

A utility function represents an individual’s preference ordering over a set of goods or outcomes, assigning a numerical value (utility) to each.

9.18.2 Types of Utility Functions

• Linear Utility Function:

  • $U(x) = ax + b$

  • Implies constant marginal utility.

• Concave Utility Function:

  • $U(x) = \sqrt{x}$ or $U(x) = \ln(x)$

  • Reflects diminishing marginal utility, common in risk-averse individuals.

• Convex Utility Function:

  • $U(x) = x^2$

  • Reflects increasing marginal utility, indicating risk-seeking behavior.

9.18.3 Risk Attitudes and Utility Functions

• Risk-Averse:

  • Prefers certain outcomes over gambles with the same expected value.

  • Utility Function: Concave (diminishing marginal utility).

• Risk-Neutral:

  • Indifferent between certain outcomes and gambles with the same expected value.

  • Utility Function: Linear.

• Risk-Seeking:

  • Prefers gambles over certain outcomes with the same expected value.

  • Utility Function: Convex (increasing marginal utility).

9.18.4 Certainty Equivalent and Risk Premium

• Certainty Equivalent (CE):

  • The guaranteed amount an individual considers equally desirable as a risky gamble.

• Risk Premium (RP):

  • The difference between the expected value of a gamble and its certainty equivalent.

  • $RP = E(X) - CE$

9.18.5 Example Calculation

Gamble:

• 50% chance to win $100.

• 50% chance to win $0.

• Expected Value (EV): $0.5 \times 100 + 0.5 \times 0 =$50$

Utility Function:

• Assume $U(x) = \sqrt{x}$

Compute Expected Utility:

• $EU = 0.5 \times U(100) + 0.5 \times U(0) = 0.5 \times 10 + 0.5 \times 0 = 5$

Find Certainty Equivalent:

• Solve $U(CE) = EU$

• $\sqrt{CE} = 5$

• $CE = 25$

Risk Premium:

• $RP = EV - CE =$50 - $25 =$25$

Interpretation:

• The individual is risk-averse and would accept $25 guaranteed over the gamble.

9.19 Game Theory Applications in Service_Design

9.19.1 Introduction to Game Theory

Definition:

Game Theory is a mathematical framework for analyzing strategic interactions among rational decision-makers, where the outcome for each participant depends on the choices of all.

9.19.2 Key Concepts in Game Theory

• Players: The decision-makers in the game.

• Strategies: The possible actions each player can take.

• Payoffs: The outcomes associated with each combination of strategies.

• Equilibrium: A state where no player can benefit by unilaterally changing their strategy (e.g., Nash Equilibrium).

9.19.3 Types of Games

• Cooperative vs. Non-Cooperative:

  • Cooperative: Players can form binding agreements.

  • Non-Cooperative: No binding agreements; focus on individual strategies.

• Zero-Sum vs. Non-Zero-Sum:

  • Zero-Sum: One player’s gain is another’s loss.

  • Non-Zero-Sum: All players can gain or lose together.

• Simultaneous vs. Sequential:

  • Simultaneous: Players make decisions at the same time.

  • Sequential: Players make decisions one after another.

9.19.4 Nash Equilibrium

Definition:

A Nash Equilibrium occurs when each player’s strategy is optimal, given the strategies of all other players. No player has anything to gain by changing only their own strategy unilaterally.

9.19.5 Application in Service Interactions

• Pricing Strategies:

  • Competitors setting prices in response to each other.

• Service Quality Levels:

  • Deciding on investment in service improvements based on competitor actions.

• Promotion and Advertising:

  • Allocation of marketing resources in a competitive environment.

9.19.6 Example: Prisoner’s Dilemma in Service_Design

Scenario:

Two competing service providers, A and B, must decide whether to invest in a new technology (Invest) or maintain current operations (Don’t Invest).

Payoff Matrix:

                      **B: Invest**           **B: Don't Invest**

A: Invest A: 3, B: 3 A: 5, B: 1

A: Don’t Invest A: 1, B: 5 A: 4, B: 4

Analysis:

• If both invest, they share the market (Payoffs: A: 3, B: 3).

• If one invests and the other doesn’t, the one who invests gains a competitive edge (Payoffs: A: 5, B: 1 or vice versa).

• If neither invests, they maintain the status quo (Payoffs: A: 4, B: 4).

Nash Equilibrium:

• Both players have an incentive to invest, leading to the equilibrium at (Invest, Invest), despite the higher collective payoff at (Don’t Invest, Don’t Invest).

Implications for Service_Design:

• Understanding strategic interactions can guide investment decisions.

• Collaboration or differentiation strategies might be considered to escape the dilemma.

9.20 Advanced Game Theory Concepts

9.20.1 Mixed Strategy Equilibria

• When players randomize over possible strategies to keep opponents indifferent.

9.20.2 Sequential Games and Backward Induction

• In sequential games, players make decisions in turns.

• Backward Induction: Solving the game by analyzing from the end backward.

9.20.3 Repeated Games

• Interactions that occur multiple times.

• Strategies:

  • Tit-for-Tat: Cooperate initially, then replicate the opponent’s previous action.

  • Grim Trigger: Cooperate until the opponent defects, then defect forever.

9.20.4 Evolutionary Game Theory

• Studies how strategies evolve over time based on their success.

• Applies to scenarios where agents adapt and learn.

9.21 Behavioral Game Theory

9.21.1 Incorporating Behavioral Insights

• Recognizes that players may not always act rationally.

• Accounts for factors like fairness, reciprocity, and bounded rationality.

9.21.2 Ultimatum Game

Game Description:

• Player 1: Proposes how to divide a sum of money.

• Player 2: Accepts or rejects the offer.

  • If accepted, the money is split as proposed.

  • If rejected, both players receive nothing.

Traditional Game Theory Prediction:

• Player 1 offers minimal amount.

• Player 2 accepts any non-zero offer.

Observed Behavior:

• Player 1 often offers 30-50%.

• Player 2 rejects low offers, even at a cost to themselves.

Implications:

• Fairness and social norms influence decisions.

• Service designers must consider these factors in interactions involving negotiations or resource allocations.

9.22 Applying Game Theory to Service Ecosystems

9.22.1 Platform Services

• Network Effects:

  • The value of the service increases as more users join.

• Strategies:

  • Pricing models to attract users (e.g., freemium models).

  • Partnering with complementors to enhance the ecosystem.

9.22.2 Competition and Collaboration

• Coopetition:

  • Competitors collaborate in certain areas while competing in others.

  • Example: Airlines forming alliances for code-sharing.

9.22.3 Regulatory Considerations

• Antitrust laws and regulations may impact strategic decisions.

9.23 Decision-Making Under Uncertainty

9.23.1 Ambiguity Aversion

• Preference for known risks over unknown risks.

9.23.2 Maximin and Maximax Strategies

• Maximin: Choose the option with the best worst-case outcome.

• Maximax: Choose the option with the best possible outcome.

9.23.3 Minimax Regret

• Minimizing the maximum regret one might experience.

Regret Table:

• Calculate regret for each option under different states of the world.

• Choose the option with the smallest maximum regret.

9.24 Incorporating Decision-Making Models into Service_Design

9.24.1 Customer Decision Journeys

• Map out the steps customers take in making decisions.

• Identify touchpoints where decision-making models can inform design.

9.24.2 Personalization and Customization

• Use data analytics to tailor services to individual preferences and decision-making styles.

9.24.3 Designing for Trust and Fairness

• Build mechanisms that foster trust, recognizing that perceptions of fairness impact user decisions.

9.24.4 Adaptive Systems

• Implement systems that learn and adapt to user behaviors, optimizing service delivery.

9.25 Ethical Considerations in Decision-Making Models

9.25.1 Transparency

• Be clear about how user data is used in decision-making processes.

9.25.2 Avoiding Manipulation

• Do not exploit cognitive biases to the detriment of users.

9.25.3 Informed Consent

• Ensure users are aware of and agree to data-driven personalization.

9.26 Conclusion of Part 3

In Part 3, we’ve examined Decision-Making Models such as Prospect Theory, Utility Functions, and Game Theory, and their applications in service design. By understanding these models, service designers can create strategies that align with how users actually make decisions, considering factors like risk preferences, perceptions of gains and losses, and strategic interactions with other market players.

Key Takeaways:

• Traditional models like Expected Utility Theory have limitations in predicting real-world behavior.

• Prospect Theory provides a more accurate representation by accounting for loss aversion and reference points.

• Utility functions help in understanding individual risk preferences.

• Game Theory analyzes strategic interactions, crucial in competitive service environments.

• Ethical considerations are essential when applying these models to ensure user trust and fairness.

In the next part, we will explore Mathematical Models that provide quantitative representations of consumer behavior and the use of Econometric Models in service design.

End of Part 3