BigDeal by Codie Sanchez · Published 2026-01-14

Charisma Expert: 5 Ways To Project Confidence (Even When You Don’t Feel It) | Dr. Shadé Zahrai

privateReview: unreviewedConfidence: high
confidenceself-doubtneurosciencecharismaimposter-syndromecalibrationaffective-labelingresponse-latencyguardrailsagent-architectureevaluationprompt-engineering
Open on YouTube ↗

Summary

Overview

  • Speaker: Dr. Shadé Zahrai
  • Channel: BigDeal by Codie Sanchez
  • Main topic: Building confidence, defeating self-doubt, and projecting charismatic presence through science-backed behavioral frameworks
  • Purpose: To provide science-backed strategies and actionable frameworks for rewiring self-doubt, projecting confidence, and building true professional presence. Dr. Shadé Zahrai, a Harvard-trained leadership coach, explains why waiting to feel confident is a flawed goal and how confidence actually follows action. She introduces frameworks for overcoming self-doubt, defeating the four drivers of imposter syndrome, and utilizing non-verbal cues, posture, and verbal framing to project magnetic charisma and competence.

Topic Map

The Confidence Fallacy: Action Precedes Confidence

  • Explanation: Waiting to feel confident or ready is a trap. Confidence is the output and result of taking action, building skills, and gathering evidence of competence.
  • Key claims:
    • Waiting to feel confident is the wrong goal.
    • Confidence comes after taking action.
    • Self-doubt stems from expectation bias and negative internal perspectives.
  • Examples:
    • Waiting to feel ready leads to endless waiting because the internal state never arrives.
  • Terminology:
    • expectation bias
    • self-efficacy
    • competence
    • evidence
  • Why it matters: Shifting the mindset from waiting for feelings to taking action breaks paralysis and accelerates personal and professional growth.

The Four Drivers of Self-Doubt

  • Explanation: Self-doubt is not a single giant blob of worry; it is driven by four distinct personality traits and psychological attributes: self-acceptance, agency, autonomy, and emotional stability.
  • Key claims:
    • Self-doubt originates from four main sources: self-acceptance, agency, autonomy (locus of control), and emotional intelligence.
    • If any of these four traits are weak, doubt sneaks in.
    • Personality traits and behaviors can be consciously managed and changed.
  • Examples:
    • An external locus of control leads people to believe life is unfair and they have no power, fueling self-doubt.
  • Terminology:
    • self-acceptance
    • agency
    • autonomy
    • locus of control
    • emotional intelligence
    • affective labeling
  • Why it matters: Diagnosing which specific driver of self-doubt is active allows individuals to target and strengthen that exact psychological area.

Projecting Charisma and Magnetic Presence

  • Explanation: Charisma is split into captivating charisma and magnetic charisma. Magnetic presence comes down to being interested rather than interesting, avoiding one-upmanship, and mastering non-verbal cues.
  • Key claims:
    • The best thing you can do to be charismatic is to be genuinely interested in others.
    • Avoid one-upmanship, which taps into a rewarding part of the brain.
    • Posture, hand gestures, smiling, and vocal pace directly influence perceived confidence and competence.
  • Examples:
    • Smiling with a warm, upturned face and maintaining appropriate eye contact builds instant rapport and perceived approachability.
  • Terminology:
    • captivating charisma
    • magnetic charisma
    • one-upmanship
    • posture
    • vocal tone
    • pausing (the 3SP method)
  • Why it matters: Mastering non-verbal and conversational cues ensures that competent individuals are also perceived as confident and approachable.

Key Points

Confidence is a lagging indicator of competence

  • Explanation: Confidence does not precede action; it is generated by taking action, building skills, and gathering evidence that you can succeed.
  • Evidence: Psychological research on goal identity and self-efficacy.
  • Practical implication: Stop waiting to feel ready; take small actions to build evidence and self-trust.

Affective labeling neutralizes emotional hijacking

  • Explanation: Instead of becoming an emotion (e.g., saying 'I am anxious'), label it objectively (e.g., 'I am noticing I am feeling anxious') to reengage the prefrontal cortex.
  • Evidence: Neuroscience studies on emotional granularity and prefrontal cortex activation.
  • Practical implication: Use precise language when describing emotions to reduce their physiological and psychological derailment.

The 3SP method enhances executive presence

  • Explanation: Pausing for two to three seconds before answering a question allows you to calm your spinning mind, control your vocal register, and project higher perceived status.
  • Evidence: Behavioral analysis of highly charismatic and magnetic speakers like Mel Robbins.
  • Practical implication: Practice intentional pauses during conversations and presentations instead of rushing to fill silence.

Frameworks, Models & Processes

The Four Drivers of Self-Doubt

  • How it works: Identifies the four root psychological traits that give rise to self-doubt, allowing targeted remediation.
  • Components:
    • Self-acceptance (worthiness and handling feedback)
    • Agency (perceived capability and overcoming imposter syndrome)
    • Autonomy (internal locus of control)
    • Emotional intelligence (managing emotional states)
  • When to use: When diagnosing the root cause of personal hesitation, fear of failure, or imposter syndrome.

The CUT Method

  • How it works: A tactical framework for dealing with manipulative behaviors by separating people from their actions and reframing self-doubt.
  • Components:
    • Cut the emotion (reframing how you interpret others' actions)
    • Unphased appearance (maintaining composure)
    • Turn off engagement (avoiding rumination and excessive mental energy expenditure)
  • When to use: When interacting with toxic or manipulative individuals in professional or personal environments.

Examples & Case Studies

Robert Robert's 1970s Dartmouth study on expectation bias gave participants fake facial scars before conversations with strangers.

  • Illustrates: Expectation bias and how internal beliefs shape external experiences.
  • Lesson: When people expect to be judged or treated poorly, they act tense and interpret neutral interactions as negative.

Tinker Hatfield transitioned from architecture to shoe design at Nike and designed the Air Nike Air Max 1 based on the Centre Pompidou.

  • Illustrates: Transferable skills and inversion thinking overcoming imposter syndrome.
  • Lesson: Focus on applying architectural thinking to a new medium instead of worrying about what you do not know.

Actionable Takeaways

  • Immediate:
    • Stop waiting to feel confident before taking action.
    • Use affective labeling ('I am noticing I feel...') instead of identifying with negative emotions.
    • Incorporate the 3-second pause (3SP method) before responding to questions.
  • Strategic:
    • Audit your four drivers of self-doubt: self-acceptance, agency, autonomy, and emotional intelligence.
    • Build self-trust through micro-commitments and habit tracking.
    • Shift from perfectionism and self-criticism to self-compassion and growth.
  • Questions to investigate:
    • Which of the four drivers of self-doubt is currently holding me back the most?
    • How does my physical posture and non-verbal communication influence how others perceive my competence?
    • Am I celebrating hard work and difficult decisions even when the final outcome falls short?

Claims Worth Verifying

  • Confidence comes after taking action, building skills, and accumulating evidence of competence. (Psychological consensus)
  • Up to 82% of people have experienced imposter syndrome at some point in their careers. (Empirical study statistic)
  • Handwritten note-taking increases recall by 20 to 30 percent compared to digital typing. (Cognitive science research)

Notable Quotes

"Waiting to feel confident is the wrong goal completely." (at 0:00) "Confidence comes after we take action." (at 0:04) "What does neuroscience show us is the main obstacle that prevents people from reaching their full potential? It entirely comes down to what we believe about ourselves." (at 2:07) "Your self-doubts are not a sign that you're broken." (at 114:01)

Compressed Summary

  • Confidence is a lagging indicator born from competence and action, not a prerequisite state.
  • Self-doubt is driven by four pillars: self-acceptance, agency, autonomy, and emotional intelligence.
  • Affective labeling switches off emotional hijacking by engaging the prefrontal cortex.
  • Magnetic presence is built through genuine curiosity, non-verbal posture, smiling, and the 3-second pause.
  • Keywords: confidence, self-doubt, neuroscience, charisma, imposter-syndrome
  • Core insight: Confidence is built through action and competence, not waiting to feel ready; by understanding the psychological drivers of self-doubt and mastering executive presence cues, anyone can project magnetic confidence.

Core insights

5
Mental Modelmedium noveltystrong evidence

Confidence is a lagging indicator: it is generated by taking action and accumulating evidence, not by waiting to feel ready. Treating confidence as a precondition creates paralysis; treating it as an output enables progress.

Why it matters

For agentic systems, this argues against gating actions on self-assessed readiness or certainty and in favor of evidential, action-based policies that accumulate competence through iterative steps.

Generalization

An AI agent's competence should be evaluated by observed task outcomes over time, not by its internal 'confidence'—and the agent should be designed to act under uncertainty while logging evidence.

Confidence does not precede action; it is generated by taking action, building skills, and gathering evidence that you can succeed.
Open source video
Waiting to feel confident is the wrong goal.
Open source video
Mechanismhigh noveltystrong evidence

Affective labeling—objectively naming an emotion ('I notice I feel anxious') instead of identifying with it ('I am anxious')—reengages prefrontal regulation and short-circuits emotional hijacking.

Why it matters

In agent/user state modeling, separating observation of emotional state from identity prevents state from contaminating the entire response; this is an architectural pattern for robust emotional handling.

Generalization

Prompt engineering for user-facing agents can benefit from a 'distancing' step: explicitly labeling the user's (or the agent's) emotional state as an observation, reducing identity fusion and improving stability.

Instead of becoming an emotion (e.g., saying 'I am anxious'), label it objectively (e.g., 'I am noticing I am feeling anxious') to reengage the prefrontal cortex.
Open source video
Practicehigh noveltymoderate evidence

The 3SP method—deliberately pausing 2–3 seconds before answering—calms internal noise, lowers vocal register, and raises perceived status and executive presence.

Why it matters

For conversational AI, response latency is usually treated as a cost; this inverts that assumption by modeling a short, deliberate cognitive pause as a quality-enhancing signal.

Generalization

A small, bounded 'thinking' delay before generating an answer may improve perceived trustworthiness and composure in voice and chat interfaces, at a latency cost that can be tuned.

Pausing for two to three seconds before answering a question allows you to calm your spinning mind, control your vocal register, and project higher perceived status.
Open source video
Mental Modelmedium noveltymoderate evidence

Self-doubt decomposes into four distinct drivers—self-acceptance, agency, autonomy, and emotional stability—so effective intervention requires diagnosing which specific driver is weak.

Why it matters

This is a diagnostic decomposition: rather than treating agent failure as one monolithic issue, separate failure causes into distinct dimensions (accepting feedback, capability/agency, locus of control, emotional robustness) for targeted remediation.

Generalization

Agent observability and evaluation should break down reliability into multiple orthogonal axes to avoid one-size-fits-all interventions.

Self-doubt is not a single giant blob of worry; it is driven by four distinct personality traits and psychological attributes: self-acceptance, agency, autonomy, and emotional stability.
Open source video
Diagnosing which specific driver of self-doubt is active allows individuals to target and strengthen that exact psychological area.
Open source video
Architecturehigh noveltyweak evidence

When dealing with manipulative behavior, the CUT method (Cut the emotion, Unphased appearance, Turn off engagement) prevents rumination and energy drain by deliberately withholding emotional reactivity.

Why it matters

For agents exposed to adversarial or manipulative input, this suggests a structured disengagement pathway—recognize, maintain composure, stop engaging—rather than always continuing the interaction.

Generalization

Agent safety guards can include an explicit 'turn off engagement' state for repeated manipulative inputs, limiting resource expenditure and potential harm.

Cut the emotion (reframing how you interpret others' actions), Unphased appearance (maintaining composure), Turn off engagement (avoiding rumination and excessive mental energy expenditure)
Open source video

Deep dives

4

Confidence calibration in LLM agents as a lagging indicator

Research question

Can an LLM agent's self-assessed confidence be measured as a lagging indicator of task success, and does acting under uncertainty converge to calibrated self-assessment?

Why

If confidence is generated by action, then gating on readiness is a fallacy; this deep dive tests whether iterative action and evidence logging produce better-calibrated agent confidence.

Confidence is the output and result of taking action, building skills, and gathering evidence of competence.
Open source video
Waiting to feel confident is the wrong goal.
Open source video
Source video

Affective labeling as a state-normalization technique for LLM chains

Research question

Does explicitly labeling emotional state as an observation (rather than identity-fused) reduce emotional bias and improve stability in LLM decision chains?

Why

Affective labeling reengages prefrontal regulation in humans; if the same pattern transfers to prompts, it offers a cheap intervention against emotional contagion in agents.

Instead of becoming an emotion (e.g., saying 'I am anxious'), label it objectively (e.g., 'I am noticing I am feeling anxious') to reengage the prefrontal cortex.
Open source video
Source video

Deliberate response latency as a perceived-composure signal in conversational AI

Research question

Does a bounded 2-3 second pause before LLM responses increase user ratings of trust and competence, and what is the latency/throughput tradeoff?

Why

Pause is usually treated as pure cost; this reframes it as a controllable quality signal for voice and chat interfaces.

Pausing for two to three seconds before answering a question allows you to calm your spinning mind, control your vocal register, and project higher perceived status.
Open source video
Source video

Multi-axis failure diagnosis for agent reliability inspired by the four drivers of self-doubt

Research question

Can a four-axis decomposition (self-acceptance, agency, autonomy, emotional stability) classify agent failure modes more effectively than a single reliability score?

Why

The decomposition enables targeted remediation by pinpointing which driver is weak, rather than treating all failures as one monolith.

Self-doubt is not a single giant blob of worry; it is driven by four distinct personality traits and psychological attributes: self-acceptance, agency, autonomy, and emotional stability.
Open source video
Diagnosing which specific driver of self-doubt is active allows individuals to target and strengthen that exact psychological area.
Open source video
Source video

Article ideas

4

Stop Gating Your AI on Confidence: Why Readiness Is a Fallacy

AI systems should treat confidence as an output of action, not a precondition, so product teams should design agents that act under uncertainty and accumulate evidence.

Angle

Direct engineering contradiction of refusal-based confidence gating, grounded in the lagging-indicator mental model.

Source video

The 2-Second Pause: Turning Latency Into a Trust Signal

A small deliberate pause before an AI response can raise perceived composure and authority, so latency budgets should be tuned for trust, not just speed.

Angle

Rebalances the assumption that faster is always better in conversational AI.

Source video

'I Notice I Feel Anxious': How Affective Labeling Can Immunize AI Prompts Against Emotional Contagion

Labeling emotional state as a separate observation in prompts reduces identity fusion and stabilizes LLM behavior on emotionally charged tasks.

Angle

Prompt-engineering pattern derived from a neuroscience mechanism.

Source video

From Imposter Syndrome to Agent Failures: A Four-Axis Diagnostic for Reliability

Agent reliability should be diagnosed across acceptance, agency, autonomy, and emotional-stability axes rather than reduced to one score.

Angle

Bridges a psychological taxonomy to agent observability and remediation.

Source video

Project ideas

4

LaggingConfidence

beyond-evals

An agent that acts under uncertainty and logs evidence across episodes will converge to better-calibrated self-assessed confidence than an agent gated on a readiness threshold.

Proof of concept

Implement a bandit/navigation task with two agent variants: one refuses actions below a confidence threshold, the other always acts and logs evidence; compare calibration over episodes.

Measurement

Expected calibration error (ECE) and task success rate after 100 episodes.

Source video

PausePercept

new

Adding a 2-3 second thinking pause before LLM responses increases user ratings of trust and competence by at least 0.5 on a 7-point Likert scale without reducing task completion.

Proof of concept

Run a between-subjects web experiment where response delay is randomly 0s, 2s, or 3s; participants rate perceived credibility after a Q&A task.

Measurement

Perceived trust/competence scores, task success, completion time.

Source video

AffectiveLabel

beyond-evals

Adding an affective-labeling sentence to prompts on emotionally charged tasks reduces variance and emotional bias in LLM outputs compared to identity-fused prompts.

Proof of concept

A/B test two prompt variants on emotionally charged customer-service scenarios; measure output valence and decision consistency.

Measurement

Valence variance, run-to-run consistency, and similarity to a neutral baseline.

Source video

CUTGuard

gatehouse

An explicit 'turn off engagement' state after repeated manipulative inputs reduces token usage and negative output while keeping false-positive disengagement below 5% on normal requests.

Proof of concept

Implement a stateful guardrail that after N manipulative inputs switches to short neutral responses; test on adversarial and benign datasets.

Measurement

Token usage, response toxicity, false-positive rate.

Source video

Architectural implications

4

Confidence is a lagging indicator, so gating on readiness is a fallacy.

Before

Agent systems require a confidence/readiness score before executing a task (e.g., refuse if uncertain).

After

Agents execute under uncertainty, building evidence over multiple episodes; confidence is a monitored output, not an input condition.

Consequence

Agents become more exploratory and robust, but require careful risk control to prevent dangerous actions under low confidence.

Source video

Affective labeling separates state from identity.

Before

User emotional state is passed in as part of the identity/context, leading to emotional contagion and drift.

After

Emotional state is represented as a separate labeled observation (e.g., 'user is anxious') not fused with task identity.

Consequence

More stable interactions, but requires extra labeling step that may add latency.

Source video

Pausing before responding enhances perceived composure.

Before

LLM APIs respond as fast as possible; latency is minimized to improve UX.

After

Interfaces deliberately insert a short 'thinking pause' before responses to increase perceived authority.

Consequence

Tradeoff: increased perceived trust vs. slower time-to-first-token; needs tuning.

Source video

CUT method's 'turn off engagement'.

Before

Agents are expected to always respond to user input, including abusive prompts.

After

Agents have an explicit disengagement state after repeated manipulative inputs.

Consequence

Better safety and resource usage, but risk of incorrectly ignoring valid requests.

Source video

Tradeoffs and failure modes

4

Deliberate pause / 3SP latency vs. responsiveness

Benefit

higher perceived composure and status; calmer internal state

Cost or risk

increases time-to-respond; in high-frequency agent loops, could hurt throughput

Pausing for two to three seconds before answering a question allows you to ... project higher perceived status.
Open source video
Source video

Affective labeling / explicit emotion observation

Benefit

reduces emotional hijacking by reengaging prefrontal cortex

Cost or risk

requires extra model inference or prompt tokens; risk of patronizing if over-used

label it objectively ... to reengage the prefrontal cortex.
Open source video
Source video

Turn off engagement / disengagement in adversarial cases

Benefit

prevents rumination and excessive mental energy expenditure

Cost or risk

may cause false negatives (ignoring legitimate needs) and trust breakdown

Turn off engagement (avoiding rumination and excessive mental energy expenditure)
Open source video
Source video

Evidence-based confidence vs. action threshold

Benefit

enables learning and progress under uncertainty

Cost or risk

increases risk of premature actions with insufficient evidence

Confidence does not precede action; it is generated by taking action...
Open source video
Source video

Open questions

4

Can an LLM-based agent's 'confidence' be measured as a lagging indicator of task success, and does acting under uncertainty converge to calibrated self-assessment?

Why unresolved

Summary only describes human psychology; translation to LLM self-reports is unverified.

Research direction

Benchmark calibration of model self-assessed confidence against observed success across repeated tasks.

Source video

Does explicitly labeling emotional state in prompts reduce emotional bias or improve downstream reliability?

Why unresolved

Evidence exists for humans, but not for LLM chains.

Research direction

A/B test prompt variations with and without affective-labeling sentences on emotionally charged tasks.

Source video

Does a short forced pause before LLM responses change user trust/competence perception?

Why unresolved

The summary describes human speakers, not LLM interfaces.

Research direction

Controlled user study varying response delay (0s, 2s, 3s) and measuring credibility ratings.

Source video

Can the four-driver decomposition be used to diagnose systemic failure modes in agent architectures?

Why unresolved

The taxonomy is for human personality, not software.

Research direction

Design multi-axis failure classification for agents: feedback acceptance, capability, control locus, emotional stability; test on failure datasets.

Source video

Key claims

6
causalVerification needed

Confidence is the output of taking action, not a precondition.

Evidence

Confidence is the output and result of taking action, building skills, and gathering evidence that you can succeed.

Question

Does this hold in controlled studies beyond correlational evidence?

Source video
factualVerification needed

Affective labeling reengages the prefrontal cortex.

Evidence

to reengage the prefrontal cortex

Question

Which specific neuroscience studies demonstrate this?

Source video
causalVerification needed

Pausing for 2–3 seconds before answering raises perceived status.

Evidence

Pausing for two to three seconds before answering a question allows you to ... project higher perceived status.

Question

Is there controlled behavioral evidence isolating pause duration?

Source video
factualVerification needed

Self-doubt is driven by four traits: self-acceptance, agency, autonomy, emotional stability.

Evidence

Self-doubt is not a single giant blob of worry; it is driven by four distinct personality traits and psychological attributes: self-acceptance, agency, autonomy, and emotional stability.

Question

Is this a validated psychometric taxonomy?

Source video
opinionVerification not requested

The best thing to do to be charismatic is to be genuinely interested in others.

Evidence

The best thing you can do to be charismatic is to be genuinely interested in others.

Source video
causalVerification needed

Avoiding one-upmanship taps into a rewarding part of the brain.

Evidence

Avoid one-upmanship, which taps into a rewarding part of the brain.

Question

What is the neural evidence for one-upmanship reward?

Source video

Connections

5