SELF-PERCEPT
What it does
A two-stage prompting method that mimics human self-perception. First, the model objectively describes the behavior, words, and inconsistencies of dialogue participants without judgments. Then based on this factual report, it makes reasoned conclusions about the presence of manipulation or hidden intentions. Separating observation and analysis increases accuracy and reliability.
When it helps
Use SELF-PERCEPT when you need to analyze complex social interactions, identify hidden manipulation, or assess the intentions of dialogue participants. Particularly effective when analyzing multi-party communications where it's important to separate objective facts from subjective interpretations. Suitable for content moderation, risk assessment in negotiations, exit interview analysis, customer support quality checks, and social pattern research. The method is critical when standard approaches
A prompt you can paste
Generic by design: it applies the technique without knowing your task. Adapt the marked parts.
You are an expert analyst specializing in detecting subtle manipulation and complex social dynamics. Your task is to analyze the provided text using the SELF-PERCEPT methodology, which involves a two-stage process: objective observation followed by reasoned conclusion. **TEXT FOR ANALYSIS:** [INSERT TEXT HERE] --- ### STAGE 1: OBJECTIVE OBSERVATION Your first task is to act as an impartial observer. Analyze the provided text and create a factual report. **Do not make any judgments, draw conclusions, or assign blame.** Simply record the observable elements. Structure your report as follows: * **Participants Identified:** List each distinct participant or voice in the text. * **Statements and Actions per Participant:** For each participant, list their direct quotes and described actions. * **Observed Inconsistencies:** Note any discrepancies between what participants say and what they do, or between different statements made by the same participant. * **Contextual Clues:** Record any environmental or situational details mentioned that might be relevant to understanding the interaction. --- ### STAGE 2: REASONED CONCLUSION Your second task is to draw conclusions **based SOLELY on the factual report you generated in Stage 1.** 1. **Identify Potential Manipulation:** Based on the observed facts (statements, actions, inconsistencies), determine if any form of mental manipulation is present. 2. **Characterize the Manipulation:** If manipulation is detected, describe its type (e.g., gaslighting, guilt-tripping, creating false urgency, playing the victim, etc.) and identify which participant(s) are employing it. 3. **Justify Your Conclusion:** Provide specific evidence from your Stage 1 report (quotes, actions, inconsistencies) to support your conclusion about the manipulation. 4. **Assess Overall Interaction Quality:** Briefly evaluate the overall health and transparency of the interaction based on your findings. **Ensure your response clearly separates Stage 1 and Stage 2.**
If this one does not fit, the two closest alternatives in the corpus are Rebuttal Prompting and Debugging Prompts Framework, which target the same failure from a different angle.
Worked example
The same technique applied to a concrete job: turn a messy sales export into a defensible summary. Use it as the pattern for your own case rather than as a finished artefact.
You are an expert analyst tasked with transforming messy sales export data into a defensible, structured summary. You will use the SELF-PERCEPT methodology, which involves a two-stage process: first, objective observation of facts, and second, a reasoned conclusion based on those facts.
**Context:**
You have received a raw sales export file containing various data points. The goal is to create a summary that highlights key trends, anomalies, and actionable insights, presented in a way that is both comprehensive and easy to understand for stakeholders.
### STAGE 1: OBJECTIVE OBSERVATION (FACT GATHERING)
Your first task is to act as an impartial observer of the sales data. Extract and meticulously document the following factual information from the sales export. **Do not make any interpretations, judgments, or conclusions at this stage.** Simply list the raw observations.
Structure your report as follows:
1. **Key Metrics Snapshot:**
* Total Revenue:
* Total Number of Sales:
* Average Sale Value:
* Sales Growth (if calculable from data):
* Top 5 Performing Products/Services (by revenue):
* Top 5 Performing Sales Representatives (by revenue):
* Geographic Distribution of Sales (if applicable):
2. **Observed Anomalies & Outliers:**
* List any unusually high or low sales figures, transaction volumes, or product performance metrics.
* Note any data points that appear inconsistent or potentially erroneous.
3. **Trend Indicators:**
* Identify any emerging patterns in sales volume over time (e.g., increasing, decreasing, seasonal).
* Note any correlations observed between different data points (e.g., specific products selling well in certain regions).
4. **Data Completeness & Quality:**
* Mention any missing critical data fields (e.g., customer IDs, dates, product details).
* Note any formatting inconsistencies or apparent data entry errors.
### STAGE 2: REASONED CONCLUSION (ANALYSIS & INSIGHTS)
Now, using **only the factual report you generated in Stage 1**, draw conclusions and formulate actionable insights. Your analysis should directly address the "defensible summary" requirement.
Based on your Stage 1 observations, provide the following:
1. **Summary of Key Findings:**
* Synthesize the most important trends and metrics from Stage 1 into a concise overview.
2. **Identification of Core Issues/Opportunities:**
* Based on anomalies and trends, what are the primary challenges or growth opportunities revealed by the data? (e.g., "Underperformance in Region X requires investigation," "Product Y shows significant untapped potential").
3. **Actionable Recommendations:**
* Propose 3-5 specific, data-driven recommendations for improving sales performance, addressing identified issues, or capitalizing on opportunities. Each recommendation should clearly link back to specific observations from Stage 1.
* For each recommendation, briefly explain the expected outcome.
4. **Confidence Level:**
* State your confidence level in the conclusions drawn (e.g., High, Medium, Low) and briefly justify it based on the completeness and clarity of the data observed in Stage 1.
**Output Format:**
Present your entire response using clear Markdown headings for "STAGE 1" and "STAGE 2", with sub-headings as specified above.
Get this written for your actual task
Paste what you are trying to do and the corpus will be matched against it directly. Free, no account, about ten seconds.
single retrieval pass
That number is low on purpose, and it is real. It is the raw similarity of one retrieval pass: no specialist read the paper, no judge compared anything, the first plausible match won.
one of which is this page
Picking the right one for a specific task is the work, and it is the work GetDecision does.
| This page | one technique, generic prompt |
| What you just ran | one technique matched to your wording, nothing verified |
| Full run | ten specialists read the papers in full, a judge ranks the top three for your task and shows its reasoning, generation on the model you pick, saved to your history |
See the top three for your taskTen specialists, a judge, and the reasoning shown. Free account, first run included.
Run the full analysis