Skip to main content

Outcome vs. Process Objective Functions in Social Perception

Core Idea​

One of the biggest differences between people is that they are optimizing for different underlying objective functions when interpreting the same social event.

This is not just a difference in opinion. It is a difference in what is being optimized.

Two Common Objective Functions​

1. Outcome-Oriented (Affect-Based)​

This objective function prioritizes:

  • Positive emotional outcomes
  • Enjoyment, play, and momentum
  • Improved interactions and behavior

The key question is:

Did this make the experience better?

If yes, then the event is evaluated as positive—even if there were hidden mechanisms involved.

Discovery of those mechanisms is often neutral or even positive (e.g., “well played”).

2. Process-Oriented (Narrative / Purity-Based)​

This objective function prioritizes:

  • Authenticity of the process
  • Coherence of the internal narrative
  • Awareness and consent to influencing factors

The key question is:

Was my perception formed in a “clean” or unmanipulated way?

If the answer is no, then the experience can be evaluated negatively—even if the outcome itself was positive.

Discovery of hidden mechanisms can retroactively invalidate the experience.

Same Event, Different Evaluation​

Consider a scenario where perception is influenced by a third party (e.g., social proof, pre-selection, signaling).

Two people can experience the exact same sequence of events:

  • One evaluates the result as positive because the interaction improved
  • The other evaluates the result as negative because the process was not transparent

The divergence is not about facts.
It is about what is being optimized.

Reaction to “Manipulation”​

A useful lens for understanding the difference:

Outcome-Oriented Response​

  • “That worked.”
  • “Interesting mechanism.”
  • “Respect.”

Process-Oriented Response​

  • “That was deceptive.”
  • “My perception was altered without consent.”
  • “This invalidates the experience.”

Both responses are internally consistent given their objective function.

Example: Engineered Motivation​

A concrete example appears in Silicon Valley.

Tracy, from HR, deliberately leverages Gilfoyle’s ego to increase his output. She mirrors his identity, reinforces his self-image, and engineers a situation where he becomes more productive.

At the end, she reveals what she did.

Gilfoyle’s response is not anger or resentment. It is:

“I respect your skills.”

From a process-oriented perspective, this is clearly manipulation:

  • His behavior was influenced without full awareness
  • The mechanism was intentionally constructed
  • The process was not “pure”

From an outcome-oriented perspective:

  • The result improved
  • The mechanism was effective
  • The reveal demonstrates competence

The same event produces completely different evaluations depending on the objective function.

Gilfoyle vs. Tracy

Internal vs. External Locus of Evaluation​

The distinction also maps to where evaluation is anchored:

Outcome-Oriented​

  • Internal locus of control
  • Focus on present interpretation and forward trajectory
  • Low sensitivity to how the state was reached

Process-Oriented​

  • External locus of control
  • Focus on historical path and correctness of formation
  • High sensitivity to unseen influences

This difference determines whether discovering a hidden mechanism produces:

  • curiosity, or
  • resentment

Robustness vs. Fragility​

These objective functions have different stability properties:

Outcome-Oriented​

  • Robust to new information
  • Does not collapse when mechanisms are revealed
  • Converts surprises into information

Process-Oriented​

  • Sensitive to hidden variables
  • Can retroactively invalidate prior states
  • Requires narrative consistency to maintain stability

Implication​

Many disagreements about “manipulation,” “authenticity,” or “ethics” are not actually about the same thing.

They are disagreements between people optimizing:

  • different objective functions,
  • over different variables,
  • with different penalty terms.

Until that difference is made explicit, discussions tend to talk past each other.

Closing​

When two people evaluate the same interaction differently, the first question is not:

Who is correct?

It is:

What is each person optimizing for?