Collaboration and mentoring: deciding better without giving away authorship

Why this matters now

Decision-making has never been completely individual. Even when one person signs a decision, what they know has been shaped by teachers, peers, specialists, communities, data, experience and disagreement.

In complex settings, this dependence becomes more visible. No individual holds every relevant discipline, knows every affected stakeholder or can foresee every consequence. Decision quality therefore depends partly on recognizing what is missing and bringing other perspectives into the process before acting.

Collaboration can also fail. Groups can reinforce shared bias, rush toward consensus, defer to the most confident voice or dissolve accountability into “the team.” A mentor can become so helpful that they effectively solve the problem. An expert can acquire authority beyond their actual field. Feedback can become instruction.

The real question is not whether we should decide alone or with others. It is how to build relationships in which external contribution improves judgment without destroying authorship.

That is the core of Collaboration and Mentoring: treating other people as part of decision infrastructure — sources of knowledge, challenge, questions and feedback — while preventing support from becoming substitution.

The core concept: think with others without giving away the decision

Collaboration and Mentoring describes two related but distinct ways in which other people can improve a decision.

Collaboration occurs when people actively contribute to understanding a problem, generating options, testing information, identifying trade-offs or constructing a shared response. Participants may have similar authority or different roles, but they influence the substance of the reasoning.

Mentoring is an asymmetrical relationship of guidance. A mentor usually brings more experience with a process, context or type of problem and uses that experience to strengthen another person's ability to think and act. The goal is not to replace the decision-maker, but to make their reasoning clearer, more rigorous and more self-aware.

Both depend on the same premise: thinking can improve when it encounters meaningful difference.

That difference can come from another discipline, profession, culture, generation or lived experience. It may also be much simpler: another person notices an inconsistency that has become invisible to someone who is too close to the problem.

More voices do not automatically create better decisions. Contribution adds value only when people can distinguish evidence from opinion, relevance from noise and support from interference.

The concept therefore includes five conditions: relevant difference, permission to question, translation across perspectives, verification and clear authorship.

It does not include obedience to a mentor, forced consensus, mechanical task allocation or transferring responsibility to the collective.

No one is ready alone

The Real-Life Readiness framework describes real life as plural. Other people bring disciplines, histories, cultures, cognitive styles, lived experience and forms of knowledge that change what a group can perceive, understand and build.

That is a stronger claim than simply saying teamwork matters.

Every individual decision is limited by what the person knows, what they can observe and which of their assumptions they can recognize. Some limits can be reduced through research. Others require encountering someone who sees the problem differently.

An engineer may understand technical viability. A designer may reveal a usability failure. Someone with local knowledge may identify a social constraint. A legal specialist may surface an obligation. A user may show that the proposed solution is solving a different problem from the one actually experienced.

These are not decorative perspectives. They can change the decision itself.

Relational preparedness begins when we stop treating other people's knowledge as an optional supplement and ask: what part of this problem am I unable to see alone?

Collaboration can become a verification architecture

A team is not useful only because it can divide work.

It can also make reasoning testable before it becomes action.

A proposal can be exposed to questions that one person might never ask:

• Is the evidence sufficient?
• Are we assuming something that has not been checked?
• Who is affected by this choice?
• Is there a technical, human or legal risk we have missed?
• Are we optimizing the wrong variable?
• Which alternative did we dismiss too early?

When questions come from different forms of expertise and experience, collaboration can function as distributed verification.

That does not mean the group is always right. Groups can create conformity and shared error.

The potential advantage emerges when the collaboration is designed to make disagreement possible, request evidence and allow direction to change.

Strong collaboration does not measure success by the absence of conflict. It measures whether relevant difference becomes useful information for judgment.

Disagreement can be information

In fragile teams, disagreement is often treated as a threat to cohesion.

In teams capable of judgment, disagreement can reveal that people are optimizing different criteria or that an important assumption has not yet been made explicit.

A solution may be technically strong and ethically weak. One person may prioritize speed; another, reversibility. One may trust a source; another may question whether the evidence is strong enough.

Disagreement becomes productive when it can be translated into inspectable questions.

Why are two reasonable people reaching different conclusions? What criterion is each person prioritizing? Does one person hold information the other lacks? Is the disagreement exposing a real trade-off?

The point is not to win the argument.

It is to use the difference to understand the decision more fully.

This connects directly to Agency. A decision with authorship is not one that excludes other perspectives. It is one whose author can explain which perspectives were considered, what trade-offs were accepted and why action was taken.

A mentor is not the answer

Mentoring introduces a particularly important boundary: one person supports the process without becoming the author of the solution.

When a mentor solves the problem for the team, the visible output may improve while the learning deteriorates. The project can become stronger at the same time that participants lose the opportunity to form judgment, experience consequences and understand why a choice was made.

Mars Challenge documentation defines this role clearly.

The mentor accompanies the process, facilitates the methodology, supports organization and gives clarity. The Official Rules add that mentors may ask triggering questions, guide reflection, help structure ideas already expressed by the team and promote ethical use of tools.

The restrictions are equally explicit. Mentors must not solve the challenge, provide the answer, direct team decisions, define technical solutions, develop the project or substitute for participants' creative work.

The mentor's value therefore does not come from possessing the answer that participants lack.

It comes from knowing how to help them produce better thinking without taking ownership of it.

The distinction can be stated simply:

answering for someone reduces authorship; asking well can strengthen judgment.

Questions are a mentoring technology

Not every question is good mentoring.

A rhetorical question designed to steer someone toward the mentor's predetermined answer may simply be an instruction in disguise.

Useful mentoring questions expose part of the reasoning.

“What evidence supports this claim?” separates confidence from support.

“Who could be affected?” introduces consequence.

“What alternatives did you consider?” tests whether a real choice occurred.

“What would make you change your mind?” tests rigidity.

“Which part of the solution can you not yet verify?” surfaces limits.

“If I were not here, how would you decide?” protects autonomy.

Questions can create productive friction between an idea and a decision. They slow the transition long enough for assumptions to become visible.

The value is not the sophistication of the question itself. It is the kind of reasoning the question makes necessary.

Feedback creates value when revision remains possible

Feedback is another form of external intelligence.

Every initial decision is made with incomplete information. Once an idea is presented, prototyped or tested, new evidence appears.

A user responds unexpectedly. An evaluator notices a contradiction. A specialist identifies a risk. A peer shows that the story is stronger than the evidence.

Feedback improves a decision only when revision remains possible.

If the goal is to defend the first idea at all costs, feedback becomes a threat. If the goal is to learn, feedback becomes new information.

This requires separating identity from decision. Changing a position in response to better evidence is not a loss of authorship. It is a more competent exercise of authorship.

Mentors and collaborators can help create that space, but the final revision must return to the person or team responsible for the work.

The boundary between influence and substitution

All collaboration influences. That is why we seek it.

The meaningful question is when influence starts to remove the author's ability to explain, justify and own the decision.

Healthy contribution can:

• provide information;
• ask questions;
• identify inconsistencies;
• introduce alternatives;
• help structure reasoning;
• surface consequences;
• offer feedback.

Substitution begins when another person:

• chooses for the team;
• produces the central solution;
• controls the narrative;
• uses authority to end discussion;
• develops the decisive parts of the work;
• leaves participants unable to explain the reasoning without external help.

This distinction is particularly important in education.

An excellent output may contain weak evidence of learning if the decisive intellectual work was performed by an adult.

Protecting authorship is therefore not only an ethical rule. It is also an evidence condition: it helps determine whether the judgment actually belongs to the person presenting it.

An explanatory synthesis: CONTRIBUTION → CHALLENGE → CLARIFY → VERIFY → REFLECT → DECIDE

To make the concept easier to use, this piece offers a six-movement editorial synthesis. It is not an additional official Mars Challenge methodology or a mandatory sequence.

CONTRIBUTION — other people add knowledge, experience or perspectives the decision-maker does not possess.

CHALLENGE — questions and disagreement test assumptions, options and excessive certainty.

CLARIFY — conversation and mentoring make objectives, criteria, evidence and trade-offs explicit.

VERIFY — different forms of knowledge help test claims and identify missing expertise.

REFLECT — feedback and consequences create an opportunity to revise reasoning.

DECIDE — the responsible person or team retains authorship of the choice and its consequences.

A useful diagnostic follows: did collaboration increase the person's capacity to decide, or did it increase dependence on the person providing help?

What the evidence says

The strongest conceptual source available is Real-Life Readiness: A Foundational Framework (2026).

Its relational root states that no one is prepared alone and that other people contribute disciplines, histories, cultures, lived experiences and forms of knowledge that change what a group can perceive, understand and build. The framework also treats translation, disagreement, verification and co-creation as part of preparedness.

This does not establish that every group outperforms every individual. It supports a narrower proposition: in complex situations, relevant knowledge may be distributed across people.

The Mars Challenge 2026 Official Rules provide specific first-party evidence about mentoring. They establish youth authorship and define the mentor as someone who asks triggering questions, guides reflection, helps structure ideas and supports ethical use of tools, while explicitly rejecting co-authorship and substitution.

Implementation materials reinforce the same logic: the mentor accompanies without solving and facilitates the process rather than supplying the answer.

The available evidence therefore supports the design logic and the boundary between support and substitution. It does not establish a universal causal magnitude for the effect of mentoring or collaboration on decision quality.

What we observe at Mars Challenge

Program documentation shows a consistent architecture linking collaboration, mentoring and authorship.

Teams move through strength identification, role allocation, idea dialogue, empathy, analysis, prototyping, evaluation, questions, refinement and final presentation.

The process is collaborative, yet the project belongs to the young participants.

Mentoring is active throughout the experience, but Official Rules prohibit mentors from becoming co-authors, making creative decisions or developing substantive parts of the project.

This avoids two opposite failures.

At one extreme, leaving a team completely alone can turn productive difficulty into disorientation.

At the other, an experienced adult can improve the project so efficiently that the very thing the learning experience is trying to reveal disappears: the participants' own judgment.

Mars Challenge's documented logic occupies the space between them:

enough support to sustain the process; enough distance to preserve Agency.

What this means for educators and mentors

When the goal includes judgment, the educator's role cannot be reduced to delivering answers.

That does not mean direct teaching is wrong. There are moments when essential knowledge is missing and instruction is appropriate.

The challenge is diagnosis.

What is missing here?

Knowledge? Structure? Evidence? A perspective? Confidence to act? Or the opportunity to practise deciding?

A competent mentor responds differently to each case.

They may introduce information when the participant genuinely lacks it. They may surface a contradiction, ask for evidence or suggest another person to consult.

But when the learner has enough material to make the choice, the mentor gives the decision back.

For institutions, this means mentor preparation should include more than challenge content. It should include boundaries of intervention, authorship, feedback, questioning, ethics, interdisciplinary work and how to support teams without over-directing them.

Success should not be measured only by whether mentors help teams produce stronger projects.

A second question matters:

does the team leave the process more capable of thinking, explaining and deciding without the mentor?

Related concepts

• Real-Life Readiness — relational preparedness and the CONNECT dimension.
• Agency — authorship, decision, action and responsibility.
• Human Judgment — evaluating evidence, context and trade-offs.
• Next Human Learning — learning that includes collaboration and reinvention.
• Interdisciplinary Learning — connecting specialisms and forms of knowledge.

Evidence in practice

• Mars Challenge 2026 — Official Rules: youth authorship principle and formal limits on mentor intervention.
• MarsChallenge2026 — Implementation 2026: active mentoring and the distinction between facilitating the process and giving the answer.
• Real-Life Readiness: A Foundational Framework (2026): conceptual basis for relational preparedness.

Frequently asked questions

Does collaboration mean every decision should be collective?

No. Some decisions belong to a particular person or role. Collaboration can improve evidence, context and alternatives before the choice is made while responsibility remains clear. Many people can contribute to one decision without all carrying the same form of accountability.

Should a mentor give the answer when they know the team is wrong?

It depends on risk and purpose. Safety, ethics and critical factual errors may require direct intervention. In a learning process, immediately correcting every error can remove opportunities to investigate, test and revise. Mars Challenge's documented model emphasizes questions and reflection while protecting the team's ownership of the solution.

What is the difference between a mentor and an expert?

An expert primarily contributes domain knowledge. A mentor primarily supports the learning and decision process. One person may occupy both roles, but it helps to make clear when they are providing specialist information and when they are helping participants form their own judgment.

Does disagreement always improve decisions?

No. Irrelevant disagreement or disagreement unsupported by evidence can create noise. It becomes useful when it reveals assumptions, criteria, risks or information that need examination. The objective is not maximum conflict; it is enough meaningful difference to test a decision before it closes.

How can we tell whether mentoring is creating dependency?

Ask whether the person or team can explain the decision, justify the evidence and continue without the mentor present. If the mentor has to choose, restructure or correct every significant step, the support may be replacing Agency rather than strengthening it.

Does this apply to collaboration with AI?

Yes, with an important distinction. AI can contribute information, alternatives and critique, but it does not automatically provide human diversity, lived context or accountability. The same principle remains: external contribution can expand judgment, while authorship and responsibility must stay clear.

References

Real-Life Readiness: A Foundational Framework Adelino Sousa / Mars Challenge / Virtual Educa 2026 Supplied source.

Official Rules of the Program — Mars Challenge 2026 Mars Challenge 2026 Supplied source.

MarsChallenge2026 — Implementation 2026 Mars Challenge 2026 Supplied source.

MC Comunicación v0.1 Mars Challenge 2026 Supplied source.

Explore next

1. Agency — how external contributions become a decision someone owns.
2. Human Judgment — how to evaluate information, AI and perspectives without delegating judgment.
3. Why Aren't Real Problems Organized by Disciplines? — why complex problems require different forms of expertise.

Next step

Deepen:

Explore Agency

Proof:

See Projects and Evidence

Action:

Discover how to take part