Additional Use Cases

While the Sensitivity Threshold Model (STM) ChatGPT interface is designed for accessible and concise education, the full model with unrestricted answer length allows for deeper exploration and powerful research-based applications. One of its standout features is the ability to layer prompts for more tailored, analytical responses—perfect for users who want to contrast STM with traditional science or test the model's alignment with academic research.

Below are two powerful ways to use this advanced prompting method to better understand, validate, and refine your grasp of the STM framework.

Evaluating STM vs. Established Science

STM ChatGPT Model

Full ChatGPT Model

Research Articles

You can prompt the model to answer a question in two layers:

First, give the standard scientific explanation based on peer-reviewed research.
Then, provide the STM interpretation.
Finally, quantify each component, showing what portion of the answer is grounded in mainstream science vs. STM theory.

This is valuable because it lets users distinguish hypothesis from consensus, and it helps maintain clarity around which aspects of the STM are novel, integrative theories and which are extensions of well-established scientific principles.

💡 Sample Prompt (Full Model):

"What causes schizophrenia? For each part of your answer, state what percentage is from established peer-reviewed science and what percentage is based on the Sensitivity Threshold Model (STM)."

Answer (Full Model): The causes of schizophrenia can be divided into two conceptual domains: established scientific consensus and the Sensitivity Threshold Model (STM). Here’s a breakdown of each key component with approximate percentage attribution based on current peer-reviewed science versus STM theory: 1. Genetic Vulnerability Established Science: ~90% Numerous genome-wide association studies (GWAS) have shown that schizophrenia is highly heritable, with polygenic risk involving genes regulating synaptic function, immune response, and neurodevelopment. STM: ~10% STM acknowledges genetic sensitivity but interprets it as contributing to a lower threshold for environmental load tolerance rather than being a direct cause. 2. Neurodevelopmental Disruption Established Science: ~80% Epidemiological and neuroimaging studies support prenatal and early-life disruptions (e.g., maternal infection, birth complications) that alter brain structure and function. STM: ~20% STM integrates this by proposing that these developmental changes set a lower cognitive and sensory processing threshold, increasing vulnerability to overload. 3. Neurochemical Dysregulation (Dopamine Hypothesis) Established Science: ~85% Excess dopaminergic activity, especially in mesolimbic pathways, is robustly linked to positive symptoms of schizophrenia (e.g., hallucinations). STM: ~15% STM reframes this dysregulation as a downstream effect of sustained cognitive and sensory overload in sensitive individuals rather than a primary cause. 4. Environmental Stress and Trauma Established Science: ~70% High urbanicity, childhood trauma, socioeconomic adversity, and migrant status are consistent environmental risk factors. STM: ~30% STM elevates these as primary causal agents in triggering overload in biologically sensitive individuals, rather than simply moderators of genetic risk. 5. Sensory Processing Sensitivity and Cognitive Overload Established Science: ~15% The Highly Sensitive Person (HSP) trait and differential susceptibility theory are supported but not widely integrated into schizophrenia models. STM: ~85% This is central to STM. It posits that schizophrenia results from sustained cognitive overload in individuals with low processing thresholds due to heightened sensitivity across sensory, emotional, and cognitive domains. 6. Neuroinflammation and Immune Dysregulation Established Science: ~60% Evidence shows elevated inflammatory markers (e.g., IL-6, TNF-alpha) and links with maternal immune activation and autoimmune comorbidity. STM: ~40% STM incorporates this as part of the cascade following threshold breach — a failure to regulate inflammation due to system overload, not immune error per se. 7. Predictive Coding and Signal Processing Errors Established Science: ~40% The predictive coding model explains psychosis as misweighted priors and aberrant salience. STM: ~60% STM can integrate this by interpreting overloaded systems as prone to prediction errors and sensory misattributions due to impaired filtering, aligning with information-theoretic overload models. 8. Antipsychotic Response and Treatment Implications Established Science: ~80% D2 receptor antagonists mitigate positive symptoms but often fail on negative/cognitive symptoms. STM: ~20% STM views antipsychotics as pharmacological dampeners of sensitivity — reducing sensory input and thus cognitive load — consistent with the model's stress-load dynamics. This hybrid view acknowledges the strong empirical grounding of biological and environmental models, while STM offers a systems-level reinterpretation that unifies these components under the principle of sensitivity-triggered overload.

Research Paper Summary and STM Analysis

This use case allows you to upload a research paper and ask the model to:

Summarize the paper clearly.
Analyze whether its findings support, contradict, or are neutral toward the STM.
Optionally request citations from STM-aligned sources or the paper itself that map onto STM principles.

This is especially useful for researchers, students, or clinicians seeking to evaluate the real-world scientific fit of the STM in the current research landscape.

💡 Sample Prompt:

"Please summarize this research paper, then evaluate whether it supports or contradicts the Sensitivity Threshold Model. Identify any parts of the paper that align with STM, and include citations that support your analysis."

Real World Example for a Recent Published Article
Article: Leveraging computational linguistics and machine learning for detection of ultra-high risk of mental health disorders in youths
PDF for Download for Model Submission

Summary of the Paper

Title: Leveraging Computational Linguistics and Machine Learning for Detection of Ultra-High Risk of Mental Health Disorders in Youths

Authors: (names not provided in the preview)

Published in: Frontiers in Psychiatry, 2022

Objective:

To investigate the feasibility of using machine learning (ML) and computational linguistics on social media posts to identify youths at ultra-high risk (UHR) for developing mental health disorders, including psychosis.

Methods:

Data Collection: Posts from Reddit subforums related to mental health and general youth communities.
Annotation: Human annotators labeled posts as UHR or non-UHR based on criteria such as disorganized thinking, bizarre beliefs, or social withdrawal.
Modeling Techniques:
- Natural Language Processing (NLP) features (e.g., LIWC, POS tagging).
- Machine Learning classifiers (Random Forests, SVMs, XGBoost).
Evaluation Metrics: Accuracy, precision, recall, and F1-score for identifying UHR youth.

Key Findings:

Models achieved up to 83% accuracy in distinguishing UHR posts.
Linguistic markers of risk included:
- First-person pronouns (linked to internal focus)
- Negative emotion words
- Disorganized language patterns
Classifiers using both linguistic and metadata features (e.g., post frequency, subreddit type) performed best.

Conclusions:

Social media language contains signals of mental health risk.
ML can augment early identification of youth at risk of psychosis and other mental disorders.
Further validation is needed, especially with clinical interviews and longitudinal follow-up.

Evaluation Against the Sensitivity Threshold Model (STM)

Does it Support STM?

Yes, it supports STM indirectly. The findings do not contradict STM and offer empirical reinforcement for several of its propositions.

Areas of Convergence

Cognitive and Emotional Markers of Overload:
- STM posits that early signs of psychosis emerge from cognitive overload in sensitive individuals.
- The paper identifies language disorganization and negative emotion as early markers — consistent with cognitive strain and emotional overload in STM .
Internal Focus and Self-Referential Language:
- STM emphasizes that sensory and emotional hypersensitivity may lead to inward cognitive focus under stress.
- Increased use of first-person pronouns, as reported in the paper, suggests internal preoccupation and reduced environmental anchoring — aligning with STM’s interpretation of stress-induced withdrawal and perceptual distortion .
Stress and Predictive Disruption:
- STM frames psychosis as a breakdown in predictive filtering under overload.
- Disorganized language could reflect impaired predictive modeling and semantic control, matching this conceptualization .
Digital Environments as Load Amplifiers:
- Although not central to the paper, the use of online posts hints at modern stressors. STM highlights urban digital environments as accelerants of overload in sensitive youth.

Neutral or Divergent Aspects

Focus on Detection Over Mechanism:
- The study is diagnostic and agnostic about causality. It does not propose a mechanism for why linguistic markers arise, whereas STM offers a detailed overload-based pathogenesis.
Lack of Individual Sensitivity Measures:
- STM requires trait sensitivity as a moderating variable. The study does not assess individual differences in sensory processing sensitivity, though these could underlie the detected linguistic features.

Conclusion

This paper supports the Sensitivity Threshold Model by offering empirical evidence that aligns with STM’s theory of early cognitive-emotional disruption in vulnerable individuals. It confirms that linguistic signs of internalization, stress, and disorganization precede diagnosable disorders — phenomena STM attributes to cognitive overload in sensitive systems. Incorporating trait sensitivity measures in future versions of this research could deepen its alignment with STM and further validate its mechanistic model.