Research methodology
The research behind Dehurdle's design.
The learning and behavioral research that shaped how Dehurdle works, and what that research does and doesn't show about Dehurdle.
The core idea
Knowing how a hard conversation should go is different from handling it under pressure. Dehurdle adds frequent, low-stakes spoken practice with AI personas that push back. Each simulation scores 5 competencies from a library of 200.
The frameworks on this page inspired how Dehurdle is designed. They weren't developed for or tested on Dehurdle, so they don't on their own show that Dehurdle improves performance.
Traditional LMS model
Passive consumption
Videos, slides, and quizzes.
No spoken practice
Dehurdle
Spoken practice
Live spoken conversations with AI personas that push back, followed by a scorecard with evidence from the transcript.
Practice out loud
Learning research that shaped the design
Deliberate practice
K. Anders Ericsson
What it says:
Ericsson's research on expert performance, starting with studies of musicians, found that experts improve through deliberate practice: effortful practice on specific weaknesses, with clear goals and immediate feedback, repeated over a long time. Experience alone isn't enough.
How it shaped Dehurdle
Each simulation focuses on one situation, such as a price objection, a difficult performance review, or a client escalation, and ends with a scorecard. You can repeat the same scenario to work on what you missed.
Stress inoculation training
Donald Meichenbaum
What it says:
A cognitive-behavioral approach from clinical psychology with three phases: understanding how stress affects you, learning and rehearsing coping skills, and applying those skills in gradually more demanding situations.
How it shaped Dehurdle
Personas range from warm and collaborative to skeptical, defensive, or aggressive, so you can practice against more pressure as you improve. Dehurdle is a practice tool, not a clinical program.
Cognitive load theory
John Sweller
What it says:
Working memory can handle only a few new pieces of information at once. Sweller argued that instruction should cut load that doesn't help learning, so attention goes to what is being learned.
How it shaped Dehurdle
Each simulation scores 5 competencies, such as objection handling, ROI articulation, or boundary enforcement, rather than a long checklist, so feedback points to a few specific behaviors.
Negotiation, conflict, and culture frameworks
Principled negotiation
Fisher, Ury and Patton, Getting to Yes (Harvard Negotiation Project)
What it says:
Four principles: separate the people from the problem, focus on interests rather than positions, invent options for mutual gain, and insist on objective criteria.
How it shaped Dehurdle
Scenarios can give a persona constraints you have to uncover, such as a budget freeze. Negotiation scenarios can score competencies based on these ideas, such as interest-based bargaining, BATNA awareness, and win-win framing.
Thomas-Kilmann conflict modes
Kenneth Thomas and Ralph Kilmann
What it says:
Describes five ways people handle conflict, based on how assertive and how cooperative they are: competing, collaborating, compromising, avoiding, and accommodating.
How it shaped Dehurdle
Dehurdle doesn't use the Thomas-Kilmann assessment. Its persona traits, such as aggressive, diplomatic, submissive, or warm and collaborative, cover a similar range of styles, so you can practice with people who handle conflict differently.
Cross-cultural communication
Geert Hofstede and Edward T. Hall
What it says:
Hofstede compared national cultures on dimensions such as power distance, individualism, and uncertainty avoidance. Hall described high-context cultures, where much of the meaning is implicit, and low-context cultures, where it is stated directly.
How it shaped Dehurdle
A catalog of 122 countries gives each persona an English accent, a short note on business etiquette, and typical phrases. For example, a Japanese persona tends to disagree indirectly. These are generalizations, and real people vary.
How scoring works
- Your speech
- Voice analysis
- Transcript
- 5 scored competencies with evidence
Voice analysis measures pace and pauses. An AI model scores the transcript.
The 200-competency framework
Dehurdle uses a library of 200 competencies, each with a written definition, grouped into five pillars. Each simulation scores 5 of them. Examples from each pillar:
- 01
Discovery
Active listening, Question framing, Paraphrasing
- 02
Value
Value articulation, ROI articulation, Conciseness
- 03
Negotiation and objections
Objection handling, Framing and reframing, Boundary enforcement
- 04
Closing
Deal closing, Stakeholder alignment, Summarization
- 05
Delivery
Pacing control, Empathy, De-escalation
Evidence from the transcript
For each competency, the AI model quotes what you said, or summarizes the relevant turn when there's no single quote, so you and your manager can see what the rating was based on. Scores are AI-generated and can be wrong. Anyone can ask for a human review of a scorecard.
Where Dehurdle fits in the Kirkpatrick model
Donald Kirkpatrick's model describes four levels for evaluating training. Dehurdle scorecards mostly inform Level 2.
| Kirkpatrick level | Core question | Where Dehurdle fits |
|---|---|---|
| Level 1: Reaction | Did participants find the training useful and engaging? | Usually a post-training survey. Dehurdle scorecards don't measure this. |
| Level 2: Learning | Did they gain the intended knowledge and skills? | Closest fit. Scorecards rate 5 competencies in each practice conversation, with evidence from the transcript. They show performance in practice, not on the job. |
| Level 3: Behavior | Do they apply what they learned on the job? | Not measured directly. Scores on repeated scenarios show change in practice over time, which you can use alongside manager observation. |
| Level 4: Results | Did the training affect business outcomes? | Measured with your own metrics, such as win rates or escalation resolution time. Dehurdle doesn't measure these. |
Frequently asked questions
Answers to common questions about the research behind Dehurdle and how scoring works.
These frameworks inspired Dehurdle's design, but they weren't tested on it. Judge Dehurdle by your own results: repeat scenarios, track scores over time, and compare them with the business metrics you already use.